Returning to a Division Centric Army

[SJPONeill]

LCol (ret) Jim Storr puts for a convincing argument of Divisional command in his book. He cites some operational analysis (on your favorite war, not the "punching bag" wars) of over 200 battles that concludes that the practical span of command for commanders is actually quite low - 1.7 subordinates committed on average to combat. This suggests that, historically, Division commanders have put forth at most 8 companies during a majority of their actions.

So that study finds that, instead of the span of 5-7 commonly held to be manageable, actual practice is much less than that i.e. 1.7? Might that be commanders focusing on their main effort operation and leaving supporting operations to their staff to manage? Does he distinguish in the study between the span of a commander i.e. the individual, and the span of command i.e. a command headquarters and staff? Please don't misread the question as critical of the statement - I think it's a fascinating insight and am just seeking clarification for myself.

Once again, just for myself, how does that reduced span link into the 8 companies in the next sentence? Is that a reduction of the number of companies by the same reduction as for the span of command?

[Fuchs]

The emphasis was on "committed" in regard to "8 companies". Storr links it also to organization and appears to dismiss whatever the other units are doing (reserve, security..) as uninteresting.
---------------------------------------------------------

I wrote a lengthy response with an example of a typically complex action, but this is not the right place for such details. This has to suffice:


The current system is not very indicative or the actual potential for another reason; a different system would need a different doctrine and different training. It's doable.
Guderian is known as champion of armoured divisions and as influence on tank development and production. His most important contribution was another one, though: He transformed his armour commanders into commanders which were useful in his new system. He had to exorcise the fear of open flanks and slowness, for example. He had to select daring officers with the inner urge to move forward. The officer corps of 1935 wasn't suitable for the armoured division actions of 1940. Those actions would have appeared foolish and impossible in 1935 to most officers (and actually kept looking foolish & impossible up to their success in May 1940). They WERE impossible with 1935's officers.

The corps concept which I'm thinking of would require a different officer training, even if applied in Germany.

[Infanteer]

So that study finds that, instead of the span of 5-7 commonly held to be manageable, actual practice is much less than that i.e. 1.7? Might that be commanders focusing on their main effort operation and leaving supporting operations to their staff to manage? Does he distinguish in the study between the span of a commander i.e. the individual, and the span of command i.e. a command headquarters and staff? Please don't misread the question as critical of the statement - I think it's a fascinating insight and am just seeking clarification for myself.

I should be careful here. One of the key assessments that Storr makes is that combat is not fractal. Brigades are not big companies and corps are not big brigades - commanders at each level have different things that they must take into consideration during the fight. A Division Commander doesn't care about the range of a GPMG, but to a platoon commander, it's one of his primary concerns. Dead ground is everything to a company but irrelevant to a Brigade.

With this principle in mind, I must make it clear that Storr only argued about the Division. The Brigade and the Corps will have different factors and thus different requirements. I made my guess of "smaller" based on some fundamental aspects of Div command which I extrapolated to the Bde and Corps level.

The figure of 1.7 is from a Dupuy study. Also citied is an unpublished DERA study which, in looking at Division activity in WWII, showed that at no time were all nine battalions of the measured division employed at the same time. Of the 81 days the measured divisions spent in combat, 43 featured only 3 battalions employed. Thus half the time divisions employed only 1/3 of their strength to defeat the enemy. Looking further at this data, divisions only employed a majority of their forces 1/3 of the time. Additional data from Suez and the Gulf further support this view. Fuchs mentioned reserves and rear duties - do these duties routinely use up over half a division's strength, especially when it is committed to frontline activity? These studies seem - at least to me - to give some concrete evidence of what the span of command actually is.

The so what - if 6-10 maneuver companies (and 2-4 battalions and 1-2 brigades) are all that a division commander can realistically employ at once, than the division should be designed around sustaining 6-10 maneuver companies in combat while, at the same time, making the organization as nimble and agile in combat. 6-10 maneuver companies do not likely require an additional 14-17 companies in reserve.

What does this mean for the Brigade and the Corps? I dunno - but the methods behind identifying the above structure (span of command, movement times, relation of space and time, etc, etc) can probably give us a good idea.

[Kiwigrunt]

Infanteer, I’ll tread carefully because I may be reading you completely wrong. Like my fellow Kiwi, I’m really just trying to get my head around this.

What does “practical span of command” actually mean? Is that the max a commander should/can be dealing with or the average he happens to deal with based on subunits in contact? If the latter is the case I should think it is perhaps a meaningless statistic.

In post 30 you say:

… the practical span of command for commanders is actually quite low - 1.7 subordinates committed on average to combat. This suggests that, historically, Division commanders have put forth at most 8 companies during a majority of their actions. Employment in DS and OIF seems to validate this observation.

Does ‘committed to combat’ suggest that the remaining units are held in reserve to the point that they are shelved till required, or just that they are not in contact?

Based on a ‘typical?’ WWII triangular infantry division, 1.7 Brigadiers may suggest about half the division. 1.7 CO’s may suggest half of those brigades. 1.7 OC’s may suggest half of those battalions. By this reasoning your estimate of 8 companies would be about right.

But this seems a bit simplistic. Firstly, 9 companies in contact could be one from each battalion, in which case the division commander has all his brigades and battalions in contact. (this is working up from your number 8 instead of down from 1.7)
Also, even if the commander holds subunits in reserve and has only one in contact most of the time that contact is made, that does not mean that he is not pulling the strings on the other units. He may be manoeuvring them and/or leapfrogging units in contact or moving one through the other etc.

This may be in line with what Fuchs said in post 35:

The emphasis was on "committed" in regard to "8 companies". Storr links it also to organization and appears to dismiss whatever the other units are doing (reserve, security..) as uninteresting.

From your previous post:

The figure of 1.7 is from a Dupuy study. Also citied is an unpublished DERA study which, in looking at Division activity in WWII, showed that at no time were all nine battalions of the measured division employed at the same time.

OK, that would counter my earlier mentioned alternative. I have not read the studies.

Looking further at this data, divisions only employed a majority of their forces 1/3 of the time.

Are these not the times that matter?

The so what - if 6-10 maneuver companies (and 2-4 battalions and 1-2 brigades) are all that a division commander can realistically employ at once, than the division should be designed around sustaining 6-10 maneuver companies in combat while, at the same time, making the organization as nimble and agile in combat. 6-10 maneuver companies do not likely require an additional 14-17 companies in reserve.

Is one third all a division can employ, or the most that a division commander likes to employ at any one time for the purpose of holding a reserve? So a question here could be: is ‘one up’ at division level realistic/sensible or just not required? Is ‘one up’ at division level really one up in the same sense that it is within companies and battalions? This in terms of the size of brigades and the relative distances involved.

So iff (wink to Tukhachevskii) 6 – 10 companies are realistically the most a division can employ at once, then should the division be reduced in size or (to support I think Ken and Fuchs) should the division be cut out of the hierarchy. If you cut down the 14 -17 companies in reserve to only a few, then what’s left is perhaps a descent size brigade.

I can think of a few extreme cases where divisions were fully employed (not including the desert). UK First Airborne in Arnhem. US 101st in Bastogne (I think).

I confuse me, I’ll stop here.

[SJPONeill]

I'm not sure that this cncept of looking at the % of a formation in combat at any one time is not a bit of a red herring...a large proportion of the force will (should) be committed to reserves at each level - I think 1/9 at each level from distant memory e.g. a section/squad at company level, a platoon at bn level, a coy at Bde level etc - that's a lot of troops when you roll them all up but I don't think it's correct to say that they are necessarily 'out of combat/contact' and thus not necessarily out of the span of comamnd equation.

I'd argue that the NZ Div on Crete and at Minqar Qaim was pretty heavily committed, as was pretty much every airborne div immediately following a combat drop in WW2...

[Fuchs]

There's usually no reserve up to company level in many armies; instead, the combat troops are meant to be capable of counter-attacks without a dedicated reserve. That's more responsive.

A battalion reserve is often not much more than a platoon, if there's any at all (a Verfügungsplatoon - a platoon directly under command of the Bn Cmdr - is an enticing idea from the Cold War; it could serve as recce Plt, as couriers, as HQ guard, as traffic organizers, as CO escort and as Bn reserve).


The share of reserves grows on formation levels, unless these formations are overstretched (on the other hand some experts think that reserves are even more important the more you're overstretched).

[Fuchs]

Kiwigrunt, I'm no fan of Storr's considerations about how few companies of a force are typically involved in combat at once. He exaggerates the point. That's especially ironic as he focuses much on the human side of war in that book.

Let's say a division has never more than ten companies at once in combat. Could it b replaced by a 10 company brigade? No!
There would be no rotation, the companies would quickly be exhausted if not depleted.
There would be no reserve, and thus no good tactics.
There would be no security elements, and thus great opportunities for OPFOR.
Perfect anticipation would be required to have these ten companies at the points of action.
OPFOR could deploy in a way which would require more than ten companies to counter.


It's one of the weaker parts of his book. My reasoning in favour of brigades rests on completely different foundations.

[TAH]

I would contend that we never stopped being division centric.

We still retain Division HQs. We still deploy Division HQs for C2.

What changed was how divisions are structured/organized.

We went from "Type" divisions (Light, Armored/Heavy, Airborne, Air Assault etc) to Modular.

So what's different? The Division "Base", the types of units organic to the division not including its maneuver battalions.

These were: ADA Bn, MI Bn, Signal Bn, Engineer Bn(s), Cavalry Sqdrn, number and types of tube/rocket Artillery Bns and their controling HQs (DIVARTY), the number and type of CSS/Support Bns and their controling HQs (DISCOM), number and type of aviation Bns and their controling HQs (Division AVN Bde) as well as MP and NBC/Chemical Companies.

In many ways, the modular BCT are acknowledgement of how we task organized maneuver brigades for operations with supporting artillery, engineers and CSS units.

The issue to be solved/re-solved is how do modular divisions operate. Who plans fires for the modular division? A section on the division staff or folks over in a assigned/attached Fires Bde? Same question/issue for division level Aviation operations.

If we follow existing CSS doctrine, Sustainment Bdes are NOT just replacements for DISCOM/COSCOMs. They are supposed to be assigned/attached to a theater-level Sustainment HQs (A TSC or ESC) and provide "area support" to all units within their designated AOR. Currently none of the deployed division HQs/CDRs are allowing that to happen. They have TACON (a command versus support relationship) of their supporting Sustainment Bde.

No adequate replacement has been found/resourced for the Division Cavalry Sqdrn. For lots of reason, its NOT a Battlefield Surveillance Bde (BFSB).

We have also not solved the issue of, I need more X but not a whole modular Bde of X, where to I go to get it?

[Infanteer]

Infanteer, I’ll tread carefully because I may be reading you completely wrong. Like my fellow Kiwi, I’m really just trying to get my head around this.

I'm not sure that this cncept of looking at the % of a formation in combat at any one time is not a bit of a red herring...

Kiwigrunt, I'm no fan of Storr's considerations about how few companies of a force are typically involved in combat at once. He exaggerates the point.

I put forward the argument to address the discussion on Span of Control(Command) that arose with Fuch's proposal of the "Division-less" Force. Storr speaks to reserves and their importance while pointing out that formation reserves can get inefficent due to span of command issues. He also points to a natural adjustment to this in WWII with US and German Divisions moving to smaller formations based on more potent Brigading in the form of "Kampfgruppes" or "Combat Commands". In a sense, a more effective division focuses on looking down and improving the Brigade (if that makes sense).

He has given his take on the span of command as a human issue based off of data he presents. Other takes on the span of command(control) are presented and many seem to be based on that common "military wisdom" that puts forward an idea as fact with no objective data or information to back it up. I will give Storr's argument that smaller formations are more flexible, efficient and effective (he looks at the Div level; Corps and Bde may have different considerations and thus different factors) credit for at least trying to draw validity from the historical record.

Before we get too far in the weeds and I end up taking Storr's argument as my own, I'll say this. We all have ideas of the perfect fighting force. Whether it be by some radical change which is too extensive to discuss here or through better training and delegation, I think it is safe to say that the "perfect" concept is just that, a concept. When making decisions, I'd ere to the side of the 80% solution based off data from our imperfect past as opposed to the current theory of the day.