Category Lessons of History

So Why Are Iraqi Records Important?

Last preachy post on this subject:

The Sad Story Of The Captured Iraqi DESERT STORM Documents

So, why does it matter? Well, Iraq has been involved in three wars in recent time that are significant.

We engaged them in the 1991 Gulf War. Even though this was a truly lopsided result, we do not have good two-sided data available for this war (it may exist, but it is classified/closed). As this was our largest conventional operation since the Korean War (1950-1953), then having good two-sided data for this war would be useful. As it is, the last major conventional fight that the U.S. participated in where there is good two-sided data is in Italy through June 1944. After that, German records significantly degrade and we do not have a good collection of opposing force data for Korea, Vietnam or Iraq….although we do have it for Grenada ;).

Then we invaded and conquered Iraq in 2003. This was effectively a three-division operation that went extremely well also. As this was the last major conventional operation we conducted, it would be useful to have good data for the opposing side (hopefully we do, but again, it is classified/closed).

But, probably most significant is the Iran-Iraq War from 1980-1988. This was the largest conventional war since the early 1970s (India-Pakistan in 1971 and the Arab-Israeli War of 1973). This was also the first major war that used chemical weapons since World War I (1914-1918). Chemical weapons and nerve agents were used extensively by Iraq against the Iranians, and also brutally against Kurdish civilians. Iran did make limited used of chemical weapons in response. It is also the only extensive use of nerve agents in warfare. This is probably important to understand and analyze. We do not have records from the Iranian side, so it would be nice to have records from the Iraqi side. Hard to analyze their operations if we have records from neither side. We really have no actual operational data on the effects of chemical weapons in combat since 1918, and we have no operational data on the effects of nerve agents in combat. This could be a major shortfall.

Now, the documents destroyed, according to the article we originally cited, was only the material captured during the 1991 Gulf War, with records dating back to 1978.  Some of it (at least 60%) was preserved on digital tapes. And, perhaps we have preserved all the military records we captured in 2003. So, in fact there may be an extensive collection available (at least 725,000 pages), although it is classified or closed to researchers. Based upon our track record of record preservation from Vietnam and onward, I have reason to be concerned.

Anyhow, a related link on our chemical warfare studies (including links to articles on the Iran-Iraq War):

Survey of German WWI Records

 

P.S. There are over a dozen counties in the world that still have chemical warfare arsenals, including Syria, Iran and North Korea.

 

U.S. Records in the Gulf War (1991)

Of course, there were problems with the U.S. Army record keeping in the Gulf War (1991). There were serious problems with the U.S. Army record keeping in the Vietnam War (1965-1973), so not surprising, the problem had not been corrected, and the same problems existed 20 years later. In the Vietnam War, the 82nd Airborne Division pretty much threw away most of their records. According to Don Hakenson, Director, Center of Unit Records Research, Records Management and Declassification Agency; in the Gulf War, 86% or 87% of the battalion daily journals were not preserved (see War by Numbers, page 146).

This became a big issue when the “Gulf War Syndrome” became an issue. People became suspicious that U.S. soldiers had become exposed to hazardous materials or chemical weapons. Yet, when the Veterans Administration and others tried to figure out where the units were at the time, they found that the records no longer existed for many these units. In many cases, they could not determine where the unit or the people were during operations. Many of the records had simply been thrown out.

The Gulf War Syndrome was not a small issue. It has been estimated that 250,000 U.S. veterans were afflicted. It was a case where record keeping briefly became a major issue. Wikipedia article: https://en.wikipedia.org/wiki/Gulf_War_syndrome

Since the 1960s, there has been serious gaps in U.S. record keeping. There still was in 1998 when we conducted a survey of the subject for the U.S. Army. We have conducted no other surveys since then, but gather that corrective action has been undertaken.

U.S. Army Record Keeping

Gulf War Records

We, of course, have never examined the other sides records for the Gulf War (1990-91). We have included in our various combat databases over 20 division and battalion-level engagements from the Gulf War. These were all assembled for us by C. Curtis Johnson, former VP of HERO and author of something like eight books.

At the time he was working for a project that had collected the U.S. Gulf War records. So he had access to the U.S. records of the various engagements, as was able to assemble the U.S. side. He had to assemble the estimates of Iraqi strength and losses based upon the U.S. intelligence records and a little educated guesswork. There are real problems in using intelligence estimates to determine the other side’s strength and losses. I can point out a number of cases where loss estimates were off by an order of magnitude (I discuss this in depth in my Kursk book). Still, as we had overrun most of the units involved, taking their records at the time, then it appears that these were reasonable and the certainly the best estimates that could be made at the time. Because the records Curt was working with were classified, and our database is unclassified, he could not leave a record of how he developed these estimates. There were, of course, also problems with the U.S. records, but that is the subject of another post.

Now, our engagements could be improved upon by a careful examination of the captured Iraqi records, which is why this caught our attention:

The Sad Story Of The Captured Iraqi DESERT STORM Documents

Needless to say, this means that for all practical purposes, the 20+ engagements in our database can never be cross-checked or improved upon. It is the best that can be done.

Captured Records: Vietnam

There is a file of captured records for the Vietnam War. The Viet Cong, having political officers and a command structure, actually did keep records. The North Vietnamese Army also kept records. During the course of the war, some of these records were captured and are in a file at the National Archives. I don’t know of anyone who has used them. I did glance at the file, and there was no finders guide and nothing was translated. There did not appear to be much order to the file. I would have needed someone fluent in Vietnamese to help me (which is actually easy to find in Northern Virginia…for example General Nguyen Ngoc Loan ended up owning a Pizza restaurant in Springfield, VA).

** EDS NOTE: GRAPHIC CONTENT ** South Vietnamese Gen. Nguyen Ngoc Loan, chief of the national police, fires his pistol, shoots, executes into the head of suspected Viet Cong officer Nguyen Van Lem (also known as Bay Lop) on a Saigon street Feb. 1, 1968, early in the Tet Offensive. (AP Photo/Eddie Adams)

In the mid-1990s I did meet with Americans who had worked with the Vietnamese in trying to locate missing U.S. servicemen. They stated that the Vietnamese were very open and interested in researching and discussing the war. They felt that they would be receptive to a joint research project on Vietnam and would be willing to open their archives for us. As we had had access to the Soviet military archives since 1993, this looked like a fairly attractive next adventure for us. Unfortunately, we could not get anyone interested in funding research on insurgencies at that time. It was not something that U.S. had researched or analyzed since 1973.

Needless to say, after we got involved in insurgencies in Afghanistan and Iraq, I again floated the idea to the Army of doing a joint research project on Vietnam. They listened to me a little longer, but in the end, there was really no interest in analyzing the insurgency in Vietnam. I am not sure why. It has the virtue of being one of the few insurgencies where the insurgents kept good records. This would allow us to do analysis based upon two-sided data. There was certainly something that could be learned from this.

Of course, one of the problems with studying Vietnam is that U.S. Army record keeping at that time was grossly substandard. It was the poorest quality records from the U.S. Army that I had ever observed. The files from most of the units were very scant. Sometimes it was difficult to even determine the units strength and losses. Some divisions were missing almost all of their files (like the 82nd Airborne). For the 1st Brigade, 5th Mechanized Division on the DMZ, we could not determine the tank strength of the unit. There was no periodic strength and loss reports for armor. For the assault helicopter battalion my father commanded, there was only a battalion newspaper and few other files. You could not tell what aircraft the unit had, nor their status or strength. It was embarrassing.

We did actually flag this problem to the active duty army of the time and they ended up giving us a contract to examine the state of current U.S, army record keeping, which is  discussed in this post:

U.S. Army Record Keeping

Anyhow, this is an extended discussion of captured records originally inspired by this post:

The Sad Story Of The Captured Iraqi DESERT STORM Documents

 

Korean War Records

Not much to say about captured records in the Korea War as I have never checked on them. I assume there must be some taken from North Korean and Chinese units and they are files away somewhere. My father did capture a Chinese soldier during the Korean War.

Oddly enough, there not been much done in the world of quantitative analysis on the Korea War outside of the work that ORO (Operations Research Office) did in the 1950s. We have never done any significant work on the Korean War. In the late 1980s we did explore conducting some analysis of Korean War battalion-level combat. As part of that effort Trevor Dupuy and I went over to the National Archives at Suitland and pulled up some U.S. Army Korea War records. They appeared to be quite complete. There were a couple of French infantry battalions attached to the U.S. Division and we appear to have good strength and loss data for them also.

Later, in 1989, Trevor Dupuy arranged with China to conduct a joint research project. It was funded by OSD Net Assessment (Andy Marshall). Trevor Dupuy really wanted to do some two-sided analysis of combat with the Chinese Army in Korea, but apparently getting access to the Chinese Army records was still too sensitive at that point. So, instead, they arranged to do a joint research contract on a more general and less sensitive theme like perceptions of each sides intentions during the Korean War. But then in June 1989 the Chinese government rolled over the student protestors in Tiananmen Square with tanks. That ended all joint research projects for many years.

We never got back to trying to conduct a joint research project on combat with China. Instead in 1993, we started a research project on Kursk using Russia records.

Trevor Dupuy did mention that the Chinese informally told him that the United States often overestimated the size of the Chinese forces they were facing, and often underestimated the casualties the Chinese took. I have no idea how valid that is.

 

Anyhow, this is an extended discussion of captured records originally inspired by this post:

The Sad Story Of The Captured Iraqi DESERT STORM Documents

Captured Records: WWII

At the end of World War I, the United States made sure they had access to the German military records in their treaty and sent a research team over there to review and copy the pertinent records. At the end of World War II, we just took everything. The allies gathered together all the material, in part because of concerns over war crimes, and eventually the entire collection was shipped back to the United States.

In the 1960s, the United States decided to repatriate the records back to Germany. Before they shipped them out they decided to copy the entire collection of German World War II records and place them on microfilm. This was a massive effort done by government contractors that took several years. Like all good government contracts, towards the end, this one was behind schedule, so they were having to cut corners by choosing not to copy things that they felt were not particularly relevant. So, the originals records are now in the archives in Freiburg, Germany (a nice little town about as far away from the old East German border as you can get). There are copies on microfilm of most of the German records, sometimes in disorder, in the U.S. Archives II in College Park, Maryland (near Washington DC). The British also have microfilm copies of portions of the German records collection that they captured over at the Public Records Office in Kew (London). The UK collection is a fraction the size of the U.S. microfilm collection, and as far as I know has nothing additional in it.

But, there were some records that were not copied by the U.S., but it is not much. For example, for the Kursk Data Base, I did the German research from the U.S. record collection. There were 17 German divisions in the offensive in the south in July 1943. I made a detailed listing of the records I have reviewed and sent that list to Dr. Arthur Volz over in Germany. He then went to Frieburg and tried to locate additional material on strength and losses from those files. About the only additional material he located was the panzer regiment files from the 11th Panzer Division, which were either not in the U.S. archives or I overlooked when I did my research. That was it. Overall, the original copying effort was pretty exhaustive.

There was one major gap for a long time. For a couple of decades, many of the original German situation maps were in the U.S., but no longer accessible. There were supposed to be copied and sent to Germany, but there was a budget issue. Meanwhile one researcher was handling them so poorly that they canceled access so as to protect them. They have finally copied them and sent the originals back to Germany.

New WWII German Maps At The National Archives

There are also no real Luftwaffe files. Most of the Luftwaffe files were placed on a train and when the order came down from Hitler to destroy everything….these weenies actually obeyed the order and burned all their records. There are also major gaps in the German records after July 1944. Every six months, the German army units wrapped up their records and sent them back to their central archives. Because the war ended in May 1945, many of the records for July-December 1944 never made it back to be filed. Same for the 1945 records. This is why the QJM (Quantified Judgment Model) was originally developed from Italian Campaign Data from 1943 through June 1944.

Anyhow, this is an extended discussion of captured records originally inspired by this post and started with the discussions below.

The Sad Story Of The Captured Iraqi DESERT STORM Documents

Captured Records: World War I

Survey of German WWI Records

 

The Historical Combat Effectiveness of Lighter-Weight Armored Forces

A Stryker Infantry Carrier Vehicle-Dragoon fires 30 mm rounds during a live-fire demonstration at Aberdeen Proving Ground, Md., Aug. 16, 2017. Soldiers with 2nd Cavalry Regiment spent six weeks at Aberdeen testing and training on the new Stryker vehicle and a remote Javelin system, which are expected to head to Germany early next year for additional user testing. (Photo Credit: Sean Kimmons)

In 2001, The Dupuy Institute conducted a study for the U.S. Army Center for Army Analysis (CAA) on the historical effectiveness of lighter-weight armored forces. At the time, the Army had developed a requirement for an Interim Armored Vehicle (IAV), lighter and more deployable than existing M1 Abrams Main Battle Tank and the M2 Bradley Infantry Fighting Vehicle, to form the backbone of the future “Objective Force.” This program would result in development of the Stryker Infantry Fighting Vehicle.

CAA initiated the TDI study at the request of Walter W. “Don” Hollis, then the Deputy Undersecretary of the Army for Operations Research (a position that was eliminated in 2006.) TDI completed and submitted “The Historical Combat Effectiveness of Lighter-Weight Armored Forces” to CAA in August 2001. It examined the effectiveness of light and medium-weight armored forces in six scenarios:

  • Conventional conflicts against an armor supported or armor heavy force.
  • Emergency insertions against an armor supported or armor heavy force.
  • Conventional conflict against a primarily infantry force (as one might encounter in sub-Saharan Africa).
  • Emergency insertion against a primarily infantry force.
  • A small to medium insurgency (includes an insurgency that develops during a peacekeeping operation).
  • A peacekeeping operation or similar Operation Other Than War (OOTW) that has some potential for violence.

The historical data the study drew upon came from 146 cases of small-scale contingency operations; U.S. involvement in Vietnam; German counterinsurgency operations in the Balkans, 1941-1945; the Philippines Campaign, 1941-42; the Normandy Campaign, 1944; the Korean War 1950-51; the Persian Gulf War, 1990-91; and U.S. and European experiences with light and medium-weight armor in World War II.

The major conclusions of the study were:

Small Scale Contingency Operations (SSCOs)

  1. Implications for the Interim Armored Vehicle (IAV) Family of Vehicles. It would appear that existing systems (M-2 and M-3 Bradley and M-113) can fulfill most requirements. Current plans to develop an advanced LAV-type vehicle may cover almost all other shortfalls. Mine protection is a design feature that should be emphasized.
  2. Implications for the Interim Brigade Combat Team (IBCT). The need for armor in SSCOs that are not conventional or closely conventional in nature is limited and rarely approaches the requirements of a brigade-size armored force.

Insurgencies

  1. Implications for the Interim Armored Vehicle (IAV) Family of Vehicles. It would appear that existing systems (M-2 and M-3 Bradley and M-113) can fulfill most requirements. The armor threat in insurgencies is very limited until the later stages if the conflict transitions to conventional war. In either case, mine protection is a design feature that may be critical.
  2. Implications for the Interim Brigade Combat Team (IBCT). It is the nature of insurgencies that rapid deployment of armor is not essential. The armor threat in insurgencies is very limited until the later stages if the conflict transitions to a conventional war and rarely approaches the requirements of a brigade-size armored force.

Conventional Warfare

Conventional Conflict Against An Armor Supported Or Armor Heavy Force

  1. Implications for the Interim Armored Vehicle (IAV) Family of Vehicles. It may be expected that opposing heavy armor in a conventional armor versus armor engagement could significantly overmatch the IAV. In this case the primary requirement would be for a weapon system that would allow the IAV to defeat the enemy armor before it could engage the IAV.
  2. Implications for the Interim Brigade Combat Team (IBCT). The IBCT could substitute as an armored cavalry force in such a scenario.

Conventional Conflict Against A Primarily Infantry Force

  1. Implications for the Interim Armored Vehicle (IAV) Family of Vehicles. This appears to be little different from those conclusions found for the use of armor in SSCOs and Insurgencies.
  2. Implications for the Interim Brigade Combat Team (IBCT). The lack of a major armor threat will make the presence of armor useful.

Emergency Insertion Against An Armor Supported Or Armor Heavy Force

  1. Implications for the Interim Armored Vehicle (IAV) Family of Vehicles. It appears that the IAV may be of great use in an emergency insertion. However, the caveat regarding the threat of being overmatched by conventional heavy armor mentioned above should not be ignored. In this case the primary requirement would be for a weapon system that would allow the IAV to defeat the enemy armor before it could engage the IAV.
  2. Implications for the Interim Brigade Combat Team (IBCT). Although the theoretical utility of the IBCT in this scenario may be great it should be noted that The Dupuy Institute was only able to find one comparable case of such a deployment which resulted in actual conflict in US military history in the last 60 years (Korea, 1950). In this case the effect of pushing forward light tanks into the face of heavier enemy tanks was marginal.

Emergency Insertion Against A Primarily Infantry Force

  1. Implications for the Interim Armored Vehicle (IAV) Family of Vehicles. The lack of a major armor threat in this scenario will make the presence of any armor useful. However, The Dupuy Institute was unable to identify the existence of any such cases in the historical record.
  2. Implications for the Interim Brigade Combat Team (IBCT). The lack of a major armor threat will make the presence of any armor useful. However, The Dupuy Institute was unable to identify the existence of any such cases in the historical record.

Other Conclusions

Wheeled Vehicles

  1. There is little historical evidence one way or the other establishing whether wheels or tracks are the preferable feature of AFVs.

Vehicle Design

  1. In SSCOs access to a large-caliber main gun was useful for demolishing obstacles and buildings. This capability is not unique and could be replaced by AT missiles armed CFVs, IFVs and APCs.
  2. Any new lighter tank-like vehicle should make its gun system the highest priority, armor secondary and mobility and maneuverability tertiary.
  3. Mine protection should be emphasized. Mines were a major threat to all types of armor in many scenarios. In many SSCOs it was the major cause of armored vehicle losses.
  4. The robust carrying capacity offered by an APC over a tank is an advantage during many SSCOs.

Terrain Issues

  1. The use of armor in urban fighting, even in SSCOs, is still limited. The threat to armor from other armor in urban terrain during SSCOs is almost nonexistent. Most urban warfare armor needs, where armor basically serves as a support weapon, can be met with light armor (CFVs, IFVs, and APCs).
  2. Vehicle weight is sometimes a limiting factor in less developed areas. In all cases where this was a problem, there was not a corresponding armor threat. As such, in almost all cases, the missions and tasks of a tank can be fulfilled with other light armor (CFVs, IFVs, or APCs).
  3. The primary terrain problem is rivers and flooded areas. It would appear that in difficult terrain, especially heavily forested terrain (areas with lots of rainfall, like jungles), a robust river crossing capability is required.

Operational Factors

  1. Emergency insertions and delaying actions sometimes appear to be a good way to lose lots of armor for limited gain. This tends to come about due to terrain problems, enemy infiltration and bypassing, and the general confusion prevalent in such operations. The Army should be careful not to piecemeal assets when inserting valuable armor resources into a ‘hot’ situation. In many cases holding back and massing the armor for defense or counter-attack may be the better option.
  2. Transportability limitations have not been a major factor in the past for determining whether lighter or heavier armor were sent into a SSCO or a combat environment.

Casualty Sensitivity

  1. In a SSCO or insurgency, in most cases the weight and armor of the AFVs is not critical. As such, one would not expect any significant changes in losses regardless of the type of AFV used (MBT, medium-weight armor, or light armor). However, the perception that US forces are not equipped with the best-protected vehicle may cause some domestic political problems. The US government is very casualty sensitive during SSCOs. Furthermore, the current US main battle tank particularly impressive, and may help provide some additional intimidation in SSCOs.
  2. In any emergency insertion scenario or conventional war scenario, the use of lighter armor could result in higher US casualties and lesser combat effectiveness. This will certainly cause some domestic political problems and may impact army morale. However by the same token, light infantry forces, unsupported by easily deployable armor could present a worse situation.

Survey of German WWI Records

At one point, we did a survey of German records from World War I. This was for an exploratory effort to look at measuring the impact of chemical weapons in an urban environment. As World War I was one of the few wars with extensive use of chemical weapons, then it was natural to look there for operational use of chemical weapons. Specifically we were looking at the use of chemical weapons in villages and such, as there was little urban combat in World War I.

As discussed in my last post on this subject, there is a two-sided collection of records in the U.S. Archives for those German units that fought the Americans in 1918. As our customer was British, they wanted to work with British units. They conducted the British research, but, they needed records from the German side. Ironically, the German World War I records were destroyed by the British bombing of Potsdam in April 1945. So where to find good opposing force data for units facing the British during World War I?

Germany did not form into a nation until 1871. During World War I, there were still several independent states, and duchies inside of the Germany and some of these maintained their own armies. The kingdoms of Bavaria, Wurttemberg and Saxony, along with the Grand Duchy of Baden fielded their own armies. They raised their own units and maintained their own records. So, if they maintained their records from World War I then we could developed a two-sided database of combat between the British and Germans in those cases where the British units opposed German units from those states.

So….for practical purposes, we ended up making a “research trip” to Freiburg (German archives), Stuttgart (Wurttemberg) and then Munich (Bavaria). Sure enough, Wurttemberg had an nice collection of records for its units (a total of seven divisions during the war) and Bavaria still had a complete collection of records for its many divisions. The Bavarian Army fielded over a dozen divisions during the war.

So we ended up in Munich for several days going through their records. Their archives were located near Munich’s Olympic Park, the place of the tragic 1972 Olympics. It was in the old Bavarian Army headquarters that had been converted to an archives. After World War II, it was occupied by the Americans, and on the doors of many of the offices was still the name tags of the American NCOs and officers who last occupied those offices. The records were in great shape. The German Army just before WWII had done a complete inventory of the Bavarian records and made sure that there were complete. It was clear that when we looked into that, that many of these files had not been opened since then. Many of the files had sixty years of dust on them. The exception was the Sixth Bavarian Reserve Division, which clearly had been accessed several times recently. Adolf Hitler had served in that division in WWI.

The staff was extremely helpful. I did bring them gifts of candy for their efforts. They were  neatly wrapped in the box with plastic mice attached to the packaging. Later, they sent me this:

So we were able to establish that good German data could be assembled for those Wurttemberg and Bavarian units that the face the British. The British company that hired us determine that the British records were good for their research efforts. So the exploratory research effort was a success, but main effort was never funded because of changing priorities among their sponsors. This research was occurring while the Iraq War (2003-2011) was going on, so sometimes budget priorities would change rather suddenly.

The Iran-Iraq War (1980-1988) also made extensive used of chemical weapons. This is discussed in depth in our newsletters. See: http://www.dupuyinstitute.org/tdipub4.htm  (issues Vol 2, No. 3; Vol 2, No. 4, Vol. 3, No 1, and Vol 3, No 2). Specifically see: http://www.dupuyinstitute.org/pdf/Issue11.pdf, page 21. To date, I am not aware of any significant work done on chemical warfare based upon their records of the war.

This post is the follow-up to these two posts:

Captured Records: World War I

The Sad Story Of The Captured Iraqi DESERT STORM Documents

The 3-to-1 Rule in Histories

I was reading a book this last week, The Blitzkrieg Legend: The 1940 Campaign in the West by Karl-Heinz Frieser (originally published in German in 1996). On page 54 it states:

According to a military rule of thumb, the attack should be numerically superior to the defender at a ratio of 3:1. That ratio goes up if the defender can fight from well developed fortification, such as the Maginot Line.

This “rule” never seems to go away. Trevor Dupuy had a chapter on it in Understanding War, published in 1987. It was Chapter 4: The Three-to-One Theory of Combat. I didn’t really bother discussing the 3-to-1 rule in my book, War by Numbers: Understanding Conventional Combat. I do have a chapter on force ratios: Chapter 2: Force Ratios. In that chapter I show a number of force ratios based on history. Here is my chart from the European Theater of Operations, 1944 (page 10):

Force Ratio…………………..Result……………..Percentage of Failure………Number of Cases

0.55 to 1.01-to-1.00…………Attack Fails………………………….100……………………………………5

1.15 to 1.88-to-1.00…………Attack usually succeeds………21…………………………………..48

1.95 to 2.56-to-1.00…………Attack usually succeeds………10…………………………………..21

2.71 to 1.00 and higher….Attack advances……………………..0…………………………………..42

 

We have also done a number of blog posts on the subject (click on our category “Force Ratios”), primarily:

Trevor Dupuy and the 3-1 Rule

You will also see in that blog post another similar chart showing the odds of success at various force ratios.

Anyhow, I kind of think that people should probably quit referencing the 3-to-1 rule. It gives it far more weight and attention than it deserves.

 

The Effects Of Dispersion On Combat

[The article below is reprinted from the December 1996 edition of The International TNDM Newsletter. A revised version appears in Christopher A. Lawrence, War by Numbers: Understanding Conventional Combat (Potomac Books, 2017), Chapter 13.]

The Effects of Dispersion on Combat
by Christopher A. Lawrence

The TNDM[1] does not play dispersion. But it is clear that dispersion has continued to increase over time, and this must have some effect on combat. This effect was identified by Trevor N. Dupuy in his various writings, starting with the Evolution of Weapons and Warfare. His graph in Understanding War of the battle casualties trends over time is presented here as Figure 1. As dispersion changes over time (dramatically), one would expect the casualties would change over time. I therefore went back to the Land Warfare Database (the 605 engagement version[2]) and proceeded to look at casualties over time and dispersion from every angle that l could.

l eventually realized that l was going to need some better definition of the time periods l was measuring to, as measuring by years scattered the data, measuring by century assembled the data in too gross a manner, and measuring by war left a confusing picture due to the number of small wars with only two or three battles in them in the Land Warfare Database. I eventually defined the wars into 14 categories, so I could fit them onto one readable graph:

To give some idea of how representative the battles listed in the LWDB were for covering the period, I have included a count of the number of battles listed in Michael Clodfelter’s two-volume book Warfare and Armed Conflict, 1618-1991. In the case of WWI, WWII and later, battles tend to be defined as a divisional-level engagement, and there were literally tens of thousands of those.

I then tested my data again looking at the 14 wars that I defined:

  • Average Strength by War (Figure 2)
  • Average Losses by War (Figure 3)
  • Percent Losses Per Day By War (Figure 4)a
  • Average People Per Kilometer By War (Figure 5)
  • Losses per Kilometer of Front by War (Figure 6)
  • Strength and Losses Per Kilometer of Front By War (Figure 7)
  • Ratio of Strength and Losses per Kilometer of Front by War (Figure 8)
  • Ratio of Strength and Loses per Kilometer of Front by Century (Figure 9)

A review of average strengths over time by century and by war showed no surprises (see Figure 2). Up through around 1900, battles were easy to define: they were one- to three-day affairs between clearly defined forces at a locale. The forces had a clear left flank and right flank that was not bounded by other friendly forces. After 1900 (and in a few cases before), warfare was fought on continuous fronts

with a ‘battle’ often being a large multi-corps operation. It is no longer clearly understood what is meant by a battle, as the forces, area covered, and duration can vary widely. For the LWDB, each battle was defined as the analyst wished. ln the case of WWI, there are a lot of very large battles which drive the average battle size up. ln the cases of the WWII, there are a lot of division-level battles, which bring the average down. In the case of the Arab-Israeli Wars, there are nothing but division and brigade-level battles, which bring the average down.

The interesting point to notice is that the average attacker strength in the 16th and 17th century is lower than the average defender strength. Later it is higher. This may be due to anomalies in our data selection.

Average loses by war (see Figure 3) suffers from the same battle definition problem.

Percent losses per day (see Figure 4) is a useful comparison through the end of the 19th Century. After that, the battles get longer and the definition of a duration of the battle is up to the analyst. Note the very dear and definite downward pattern of percent loses per day from the Napoleonic Wars through the Arab-Israeli Wars. Here is a very clear indication of the effects of dispersion. It would appear that from the 1600s to the 1800s the pattern was effectively constant and level, then declines in a very systematic pattern. This partially contradicts Trevor Dupuy’s writing and graphs (see Figure 1). It does appear that after this period of decline that the percent losses per day are being set at a new, much lower plateau. Percent losses per day by war is attached.

Looking at the actual subject of the dispersion of people (measured in people per kilometer of front) remained relatively constant from 1600 through the American Civil War (see Figure 5). Trevor Dupuy defined dispersion as the number of people in a box-like area. Unfortunately, l do not know how to measure that. lean clearly identify the left and right of a unit, but it is more difficult to tell how deep it is Furthermore, density of occupation of this box is far from uniform, with a very forward bias By the same token, fire delivered into this box is also not uniform, with a very forward bias. Therefore, l am quite comfortable measuring dispersion based upon unit frontage, more so than front multiplied by depth.

Note, when comparing the Napoleonic Wars to the American Civil War that the dispersion remains about the same. Yet, if you look at the average casualties (Figure 3) and the average percent casualties per day (Figure 4), it is clear that the rate of casualty accumulation is lower in the American Civil War (this again partially contradicts Dupuy‘s writings). There is no question that with the advent of the Minié ball, allowing for rapid-fire rifled muskets, the ability to deliver accurate firepower increased.

As you will also note, the average people per linear kilometer between WWI and WWII differs by a factor of a little over 1.5 to 1. Yet the actual difference in casualties (see Figure 4) is much greater. While one can just postulate that the difference is the change in dispersion squared (basically Dupuy‘s approach), this does not seem to explain the complete difference, especially the difference between the Napoleonic Wars and the Civil War.

lnstead of discussing dispersion, we should be discussing “casualty reduction efforts.” This basically consists of three elements:

  • Dispersion (D)
  • Increased engagement ranges (R)
  • More individual use of cover and concealment (C&C).

These three factors together result in the reduced chance to hit. They are also partially interrelated, as one cannot make more individual use of cover and concealment unless one is allowed to disperse. So, therefore. The need for cover and concealment increases the desire to disperse and the process of dispersing allows one to use more cover and concealment.

Command and control are integrated into this construct as being something that allows dispersion, and dispersion creates the need for better command control. Therefore, improved command and control in this construct does not operate as a force modifier, but enables a force to disperse.

Intelligence becomes more necessary as the opposing forces use cover and concealment and the ranges of engagement increase. By the same token, improved intelligence allows you to increase the range of engagement and forces the enemy to use better concealment.

This whole construct could be represented by the diagram at the top of the next page.

Now, I may have said the obvious here, but this construct is probably provable in each individual element, and the overall outcome is measurable. Each individual connection between these boxes may also be measurable.

Therefore, to measure the effects of reduced chance to hit, one would need to measure the following formula (assuming these formulae are close to being correct):

(K * ΔD) + (K * ΔC&C) + (K * ΔR) = H

(K * ΔC2) = ΔD

(K * ΔD) = ΔC&C

(K * ΔW) + (K * ΔI) = ΔR

K = a constant
Δ = the change in….. (alias “Delta”)
D = Dispersion
C&C = Cover & Concealment
R = Engagement Range
W = Weapon’s Characteristics
H = the chance to hit
C2 = Command and control
I = Intelligence or ability to observe

Also, certain actions lead to a desire for certain technological and system improvements. This includes the effect of increased dispersion leading to a need for better C2 and increased range leading to a need for better intelligence. I am not sure these are measurable.

I have also shown in the diagram how the enemy impacts upon this. There is also an interrelated mirror image of this construct for the other side.

I am focusing on this because l really want to come up with some means of measuring the effects of a “revolution in warfare.” The last 400 years of human history have given us more revolutionary inventions impacting war than we can reasonably expect to see in the next 100 years. In particular, I would like to measure the impact of increased weapon accuracy, improved intelligence, and improved C2 on combat.

For the purposes of the TNDM, I would very specifically like to work out an attrition multiplier for battles before WWII (and theoretically after WWII) based upon reduced chance to be hit (“dispersion”). For example, Dave Bongard is currently using an attrition multiplier of 4 for his WWI engagements that he is running for the battalion-level validation data base.[3] No one can point to a piece of paper saying this is the value that should be used. Dave picked this value based upon experience and familiarity with the period.

I have also attached Average Loses per Kilometer of Front by War (see Figure 6 above), and a summary chart showing the two on the same chart (see figure 7 above).

The values from these charts are:

The TNDM sets WWII dispersion factor at 3,000 (which l gather translates into 30,000 men per square kilometer). The above data shows a linear dispersion per kilometer of 2,992 men, so this number parallels Dupuy‘s figures.

The final chart I have included is the Ratio of Strength and Losses per Kilometer of Front by War (Figure 8). Each line on the bar graph measures the average ratio of strength over casualties for either the attacker or defender. Being a ratio, unusual outcomes resulted in some really unusually high ratios. I took the liberty of taking out six

data points because they appeared unusually lop-sided. Three of these points are from the English Civil War and were way out of line with everything else. These were the three Scottish battles where you had a small group of mostly sword-armed troops defeating a “modem” army. Also, Walcourt (1689), Front Royal (1862), and Calbritto (1943) were removed. L also have included the same chart, except by century (Figure 9).
Again, one sees a consistency in results in over 300+ years of war, in this case going all the way through WWI, then sees an entirely different pattern with WWII and the Arab-Israeli Wars

A very tentative set of conclusions from all this is:

  1. Dispersion has been relatively constant and driven by factors other than firepower from 1600-1815.
  2. Since the Napoleonic Wars, units have increasingly dispersed (found ways to reduce their chance to be hit) in response to increased lethality of weapons.
  3. As a result of this increased dispersion, casualties in a given space have declined.
  4. The ratio of this decline in casualties over area have been roughly proportional to the strength over an area from 1600 through WWI. Starting with WWII, it appears that people have dispersed faster than weapons lethality, and this trend has continued.
  5. In effect, people dispersed in direct relation to increased firepower from 1815 through 1920, and then after that time dispersed faster than the increase in lethality.
  6. It appears that since WWII, people have gone back to dispersing (reducing their chance to be hit) at the same rate that firepower is increasing.
  7. Effectively, there are four patterns of casualties in modem war:

Period 1 (1600 – 1815): Period of Stability

  • Short battles
  • Short frontages
  • High attrition per day
  • Constant dispersion
  • Dispersion decreasing slightly after late 1700s
  • Attrition decreasing slightly after mid-1700s.

Period 2 (1816 – 1905): Period of Adjustment

  • Longer battles
  • Longer frontages
  • Lower attrition per day
  • Increasing dispersion
  • Dispersion increasing slightly faster than lethality

Period 3 (1912 – 1920): Period of Transition

  • Long Battles
  • Continuous Frontages
  • Lower attrition per day
  • Increasing dispersion
  • Relative lethality per kilometer similar to past, but lower
  • Dispersion increasing slightly faster than lethality

Period 4 (1937 – present): Modern Warfare

  • Long Battles
  • Continuous Frontages
  • Low Attrition per day
  • High dispersion (perhaps constant?)
  • Relatively lethality per kilometer much lower than the past
  • Dispersion increased much faster than lethality going into the period.
  • Dispersion increased at the same rate as lethality within the period.

So the question is whether warfare of the next 50 years will see a new “period of adjustment,” where the rate of dispersion (and other factors) adjusts in direct proportion to increased lethality, or will there be a significant change in the nature of war?

Note that when l use the word “dispersion” above, l often mean “reduced chance to be hit,” which consists of dispersion, increased engagement ranges, and use of cover & concealment.

One of the reasons l wandered into this subject was to see if the TNDM can be used for predicting combat before WWII. l then spent the next few days attempting to find some correlation between dispersion and casualties. Using the data on historical dispersion provided above, l created a mathematical formulation and tested that against the actual historical data points, and could not get any type of fit.

I then locked at the length of battles over time, at one-day battles, and attempted to find a pattern. I could find none. I also looked at other permutations, but did not keep a record of my attempts. I then looked through the work done by Dean Hartley (Oakridge) with the LWDB and called Paul Davis (RAND) to see if there was anyone who had found any correlation between dispersion and casualties, and they had not noted any.

It became clear to me that if there is any such correlation, it is buried so deep in the data that it cannot be found by any casual search. I suspect that I can find a mathematical correlation between weapon lethality, reduced chance to hit (including dispersion), and casualties. This would require some improvement to the data, some systematic measure of weapons lethality, and some serious regression analysis. I unfortunately cannot pursue this at this time.

Finally, for reference, l have attached two charts showing the duration of the battles in the LWDB in days (Figure 10, Duration of Battles Over Time and Figure 11, A Count of the Duration of Battles by War).

NOTES

[1] The Tactical Numerical Deterministic Model, a combat model developed by Trevor Dupuy in 1990-1991 as the follow-up to his Quantified Judgement Model. Dr. James G. Taylor and Jose Perez also contributed to the TNDM’s development.

[2] TDI’s Land Warfare Database (LWDB) was a revised version of a database created by the Historical Evaluation Research Organization (HERO) for the then-U.S. Army Concepts and Analysis Agency (now known as the U.S. Army Center for Army Analysis (CAA)) in 1984. Since the original publication of this article, TDI expanded and revised the data into a suite of databases.

[3] This matter is discussed in Christopher A. Lawrence, “The Second Test of the TNDM Battalion-Level Validations: Predicting Casualties,” The International TNDM Newsletter, April 1997, pp. 40-50.