All too often studies of Civil War command are in reality studies of Civil War commanders. We know very little about how an army’s command system works, the responsibilities of command, and how command is exercised. For example, we don’t pay much attention to logistics and staff work, information gathering and battlefield communication.
I’ve had reason to contemplate these questions as I’ve written about commanders and command. I think one of the big reasons why Gettysburg turned out as it did was that for once the Union command system, even in somewhat ad hoc form, proved superior to its Confederate counterpart. In part that was due to the initiative of brigade commanders, and, all stories about Joshua Chamberlain aside, brigade commanders in blue were essential to the outcome of that battle, as were the performances of Winfield Scott Hancock and Gouverneur K. Warren. Hancock seems to me to be a special case here, because he did not function as a corps commander during the battle (although he is often celebrated as the Army of the Potomac’s best corps commander, a ranking which might say something about the competition for that title, as Hancock’s performance in 1864 was a bit more mixed, although not all of that is due to him).
So, you tell me: what were the responsibilities of a corps commander, and who met those responsibilities best?
Have you read R.S. Jones’ “The Right Hand of Command”? He makes a pretty good case that Grant was, yet again, the leader in developing a staff organization capable of exercising command and control in the field sufficient for the forces involved.
Robert Berlin’s “United States Army World War II Corps Commanders” which can be found at the CSI website, defines the responsibilities of a corps commander as:
“…the highest-level officer who was engaged in battle at the front and who concentrated on high-level tactics. Leaving administrative matters largely at army level, be coordinated the use of-combined arms on the battlefield. He was, as one general describes him, “the last man towards the rear who directs tactical fire on the enemy. He is the commander who conducts the battle.” The corps commander was responsible for coordinating and directing the effort of the corps as a combined arms whole…” which seems reasonable in an ACW context as well.
The Civil War corps functioned as the operational headquarters at the level straddling what today is called the tactical/operational divide (what used to be referred to as “grand tactics”) but below the “strategic” level provided during the ACW by a department or army headquarters.
As such, the ideal corps commander had to combine the both offensive and defensive acumen with administrative abilities and aggresiveness that would allow his command to function both as part of a larger army-level team and as an independent force when necessary. That level of professionalism is not easy to achieve; in terms of the US forces during the ACW, the top performers come to mind as individuals who could balance the strengths of their subordinates (Lee using Jackson as the hammer and Longstreet as the anvil, for example) in operational terms while providing the trust necessary up and down the chain of command, and the clear direction and instruction necessary in an age where what is referred to as command and control was only just becoming faster than the pace of a running horse…
Looking at the field armies that were split into corps in 1862, it strikes me that McClellan actually had a decently experienced group for the time – McDowell had more experience than anyone else at the timwe commanding a multi-division force; Sumner was a regular with significant expeirence in a political-military theater (Kansas) that was a closer analogue to civil war than anything else the US Army had faced since the Revolution; Heitzelman was another senior regular and had been a division commander under McDowell in the 1st Bull Run campaign; and Keyes was a senior aide to Scott in the pre-war years (as McDowell had been) and a brigade commander under McDowell.
Franklin and Porter were a younger generation, although both were regulars and Franklin did have a brigade under McDowell. I think MacClellan’s idea of waiting until a division commander “proved” himself before forming corps was pretty questionable; with 14 divisions formed by December, 1861, his span of control was almost three times what is historically recommended as the most efficient.
McDowell’s division commanders for the 1st Bull Run campaign were Tyler, Hunter, Heintzelman, Runyon, and Miles; Patterson’s at the same time were Cadwallader, Dix, Keim, and Sanford; Banks had a division in the Shenandoah at the same time. Setting aside the militia generals, seems like Heintzelman was the best by a fair piece.
Brigade commanders in the same period included Thomas, Williams, Longenecker, Wynkoop, Negley, Abercrombie, Stone, Butterfield, Schenck, WT Sherman, Richardson, Andrew Porter, Burnside, Franklin, Wilcox, Ward, Howard, Blenker, Davies, and Stiles. Brigade commanders in McClellan’s first organization also included Kearney, WF Smith, Couch, and Graham, while his West Virginia campiagn division/brigade commanders included Rosecrans, Kelly, T.A. Morris, C. W. Hills, R.L. McCook, Cox, Benham, and J.J. Reynolds.
Interesting to consider moving some of the names around…
Have to think about it some more.
Interesting points. Of course, as you note, McClellan objected to the corps structure when it was imposed, although that may have been based as much on “who” as on “what”. I see some irony in that It seems to have been determined during 1862 on both sides of the fence that a higher organization beyond the division level was needed (Shiloh, Seven Pines, etc.) because those battles demonstrated that while the divisions could be effective fighting forces something was lacking in terms of coordination. At least on the Union side, this produced a lot of mediocrity. It worked better on the Confederate side, for awhile anyway. No doubt it had to do with Lee’s access to, and use of, Longstreet and Jackson (compared to the use of Longstreet, Ewell, Hill, etc.) but I also note the substantially larger size of the Rebel corps – which gets us to Burnside’s Grand Divisions and how that concept didn’t work.
I have to think more about this. For now, I’d observe that on the Union side corps command seems to have been a “Mediocrity Melting Pot”. Was that due to (1) something inherent about a command being intermediate between a division and an army: (2) the skill sets of the gents who got those posts; or (3) a basic organizational problem. The names of guys who come readily to mind as successful are at the brigade/division level – and some of them are the same names that were to one degree or another underwhelming at the next level. The only thing I can focus on right now is the increasingly remote ability to influence by direct command/presence – or maybe undue deference to seniority Those should also apply to the army command level as well, however. Just as an aside, I’d certainly add Hunt to your Gettysburg list in his difficult-to-categorize artillery post.
There wouldn’t happen to be out there a collection of essays in the works that addresses this subject, one that is only waiting on the last of its contributors to finally get his essay in, is there?
I knew you would chime in.
One reason the NVA did so well is that the Confederates recognized it was better have a few good Corps commanders rather than a large number of mediocre ones. So, at Gettysburg Lee had 3 Corps Commanders and 9 Divisional commanders, Meade had 7 Corps and 21 Divisional commanders. The AoP would’ve been better off with 4 Good Corps commanders and 12 Divisional Commanders. I think Hooker was an extremely good Corps commander, but was promoted above his ceiling – he also didn’t play well with others, which was a problem given the way the Union army operated.
But of those 3 corps commanders, only Longstreet really knew what he was doing. Hill was too sick to be of any resourcefulness, and Ewell, in his first battle back, was suddenly in charge of a corps and without the leadership of Jackson.
That is a good point – remember, the initiual AoTP corps organization amounted to four, each made up of three divisions, with Banks’ V Corps (2 divisions) essentially detached to operate in the Shenandoah.
The three-division organization went back to Napoleon and the Grand Army, and it strikes me that McClellan undoubtledy was using that as his model.
When the army got to the Peninsula, McClellan reformed it over the course of the campaign, essentially, into five two-division corps, which gave him two corps commanders he apppointed, but I don’t know if it really improved the command and control; seems like he could have done as well (or as poorly) in the Seven Days with Sumner, Heintzelman, and Keyes running three corps of three divisions each, plus a “fire brigade” (presumably Sykes’ division and the cavalry) as he did historically with five corps of two divisions each…
It is hard to criticize the AoTP corps commanders given the challenges inherent of doing so absent McClellan, but it seems Heintzelman and Keyes did about as well as can be expected at Williamsburg, given McClellan’s decision-making; Sumner’s record is mixed, but he should get points for the Chickahominy and generally being ready to fight whenever necessary; Franklin and Porter both did what McClellan asked them too, and Porter deserves credit for Malvern Hill.
Their opposite numbers, at Seven Pines and Seven Days, (including GW Smith, Magruder, and Holmes, along with Jackson and Longstreet) seem to be an equally mixed bag, to be honest about it.
In the west, measuring Bragg’s corps commanders against Buells’, for example, seems to present something similar…Thomas is the standout; maybe Hardee on the CSA side.
Best,
This is such a great question, which I think makes it harder to answer. Corps commanders were intermediaries between Army HQ and tactical units. As TF Smith put it, they straddle the tactical/operational divide. There tends to be a heavy emphasis on battlefield leadership but I think there should also be some thought put toward non-battle operations — movement and logistical management for example. I don’t have a clear understanding (and I think this is true of a lot of people) of the division of administrative/staff/logistical responsibilities at different levels in a civil war army, which makes it hard to judge how well these responsibilities were managed.
Jones’ book is a good source; gives a nice review of the state of the art, so to speak, in 1860 in Europe, how the US forces reached that level during the course of the war, and provides case studies of McClellan, Lee, Grant, and Sherman.
The focus is at the army level, but gives some insight into corps level, as well.
There’s a fair amount of material in various monographs over the years at the professional schools, but not a whole lot of that gets into the general civil war history side of things.
Parameters and the CSI monograph series – which are available on-line – are well worth review; Parameters focuses a lot on the “here and now” of the time, but there is a lot of good history interspersed with the (at the time) focus on the Soviets, or counterinsurgency, or whatever…
Best,
I wonder if the problems with corps command weren’t simply due to the concept of the corps itself. The corps was too big to be efficiently handled, but often too small to cover the topological area assigned. In these situations, external divisions or brigades could be attached to the corps but typically command and control was terrible (I’m thinking Port Gibson and Champion HIll).
I think Lee had it right with the idea of wing commanders, and suffered for the decision to invade the north with three corps commanders in 1863.
My working definition has been Corps is the largest combined-arms force that can operate along a single line of march. Because of the logistics and maneuver constraints the Corps is limited in its ability to support, or be supported by, other units. The problem for Army command is “march separately, fight jointly”. Thus for Corps command the problem is to operate as an independent commander until such time as his force is brought into support range of other units, and then shift from independent command to mutual support.
I think you are right on concentrating on the movement issue; vanguards, right, left, and center, and rearguards go back to the Romans, presumably earlier. The Civil War field armies that included a “reserve” corps, specifically named as such, are an echo of this, I think…but in a war that saw “legions” marching on both sides, seems appropriate.
The idea of the corps, as promulgated by Napoleon, was that it should be a combined arms team Iinfantry, cavalry, artillery) capable of fighting by itself and holding within a short (one day or less) march of the other elements of a given field army.
I think one of the problems with most of the Napoleonic precepts in 1860s North America were a) the longer range of rifles and artillery, vis a vis 50 years earlier; and b) at least in comparison to western Europe (as opposed to eastern) the relative paucity of the road net.
So, Dr. Simpson, what are you thoughts in terms of answering the questions you have posed?
I suppose one way to evaluate the 1862 corps commanders is to consider which ones made it through their service absent being dismissed by the seniors; by that measure, of the first five AoTP corps commanders (McDowell, Sumner, Heintzelman, Keyes, and Banks) only Sumner and Banks (!) were promoted above that level (grand division and the Army of the Gulf); I would suggest that only Sumner departed active service with his reputation intact, since my recollection is that after his request for reassignment but prior to his death, he was in line for a major command in the West – I expect Missouri, but does anyone know for certain?
Porter and Franklin each got AoTP corps in 1862; Porter never went any higher, but Franklin got a grand division under Burnside – who also presumably counts as an 1862 corps commander, and who got not one but two separate armt level commands during the course of the war.
The Antietam corps commanders included Hooker, Sumner, Porter, Franklin, Burnside, and Mansfield; Hooker got a grand division and then an army (and then a corps again); Mansfield was killed, and it is hard to say if his promotion was much more than a tombstone; I don’t know if Cox and Meade can really be regarded as exercising corp command.
Of the three corps commanders in the 1862 Army of Virginia (Banks, Sigel, and McDowell), it is pretty evident none were highly regarded as time went on; of the first three in the Army of the Ohio, (McCook, Crittenden, and Gilbert) none got any higher positions; in the Army (Group) of the Mississippi (Halleck’s command post-Shiloh), Thomas, Buell, Pope, and McClernand acted, essentially, as corps commanders. All four ended their service at the army level, but presumably only Thomas was well-regarded at the end of his service.
Looking back, interesting that of the US Big Five, only Grant and Thomas had responsibilities greater than at the divisional level in 1862.
Best,
Sumner had been ordered to take command of the Department of the Missouri but died before he could go.
Just listing those names is a useful exercise. In NFL terms, that looks an awful lot like a “weak draft” year – akthough it’s difficult to figure out what’s the “chicken” and what’s the “egg” here. Gilbert, by the way, should have an asterisk next to his name. His rank was well below anything which would justify corps command and he was a (very) temporary appointment due to the bizarre circumstances affecting the Army of the Ohio command in late September/early October, 1862.