McClellan Re-evaluating George McClellan

67th Tigers: http://civilwartalk.com/threads/re-evaluating-george-mcclellan.100689/page-15#post-943410 I seem to have missed any clarification of where your definition of "fit to march" and "present for duty" are the same category is from.

I know this discussion has moved on from Lee's numbers as observed by Steiner, but given that any rating of McClellan's actions being less than could be done needs to look at that, what McClellan faced still needs to be defined based on something other than the overestimates of unreliable intelligence gatherers or the underestimates of post-war Lost Cause writing.
 
67th Tigers: http://civilwartalk.com/threads/re-evaluating-george-mcclellan.100689/page-15#post-943410 I seem to have missed any clarification of where your definition of "fit to march" and "present for duty" are the same category is from.

I know this discussion has moved on from Lee's numbers as observed by Steiner, but given that any rating of McClellan's actions being less than could be done needs to look at that, what McClellan faced still needs to be defined based on something other than the overestimates of unreliable intelligence gatherers or the underestimates of post-war Lost Cause writing.

Elennsar,

I noticed that over the last month that my name and research have come up in this thread a number of times regarding my article in the Washington Post about Lee's numbers during the Antietam Campaign.

When the article was published two years ago I also linked to some supporting material at the following urls :

http://www.washingtonpost.com/wp-srv/special/artsandliving/civilwar/mcclellan-graphic/?tab=3

http://www.washingtonpost.com/wp-srv/special/artsandliving/civilwar/mcclellan-graphic/?tab=2

Did you happen to read those links before writing some of your less than complimentary comments about my ability to do research?

Sincerely,

Gene Thorp
 
Mr. Thorp:

Your first link (in your post above) asserting that we can determine Lee's starting strength by adding the campaign's causalities to the October return risks double counting anyone noted as a causality at some point during the campaign but who had returned to the ranks as of the October 10, or any men who had not been present at the start of the campaign returning to the ranks in time for the October 10 count being accidentally included as part of the campaign start count.

This is less troubling than reporting Lee's "infantry PFD" based on a return that includes two artillery battalions as well as divisional artillery (certainly it is not listed as missing, and is usually given when divisional totals are), but it is not any better than anyone else's guesstimate.

And your use of eyewitness head counts to justify claims of Lee's numbers does not address that even were those headcounts 100% accurate that they would be counting the total men marching by - a poor basis for determining Lee's present for duty as distinct from his aggregate present, and yet used as if these totals were both certainly reliable and certainly of present for duty only.

I have no doubt about your interest in documenting Lee's numbers, but the soundness of your writing I find doubtful based on that being offered as supporting your conclusion.

As for the issue of Lee having lost more men than he is generally acknowledged as losing, that is not an area I have investigated myself to any extent, and given that it's a pretty recognizable fact that both sides undercounted their own losses it would not surprise me to see that happen here as well.

But it does not address the issue of the strength of his army being too much for McClellan's own forces, which is my main concern in regards to McClellan's actions, and thus is not one I find particularly convincing as relates to the defense of McClellan's actions.

Frankly, I feel that neither this or Sears are convincing, and look to find better information in other sources when they're available.
 
"Add back" is a reasonable technique.

Newton estimates for 1st June, before Seven Pines casualties deducted = 94,813

Harsh estimates for 26th June =: 101,725 exclusive of Department of Henrico

20th July: 69,732 exclusive of Jackson (5,020 exc/ arty etc. on 14th August morning report) and Ewell's divisions (5,027 on 10th August) and Lawton's brigade (2,041 on 13th August when it officially was assigned to Ewell) = ca. 81,820

Harsh estimates on 2nd September PFD = 75,528

22nd September: 36,418 PFD exclusive of reserve artillery and cavalry (neither of which heavily engaged at Antietam). Est 771 (30th Sept) for arty and 5,761 (10th Oct) for cav = 42,950

30th September: 52,790 PFD exclusive of cavalry (reserve arty reported at 771). Est 5,761 for cav = 58,551

10th October: 64,273 PFD (reserve arty reported at 912, cavalry at 5,761)
20th October: 68,033
10th November: 70,909 exclusive of reserve artillery - add 917 res arty (20th Oct) = 71,826
20th November: 73,778 exclusive of reserve artillery - add 917 res arty = 74,695
31st December: 76,799

Graphically:

ANV%2BStrength%2B2nd%2Bhalf%2B62.png

The "add back" gives a number very close to Harsh, and adding back say the Seven Day casualties to the post Seven Days strengths gives numbers close to Harsh. When two methods converge it is very indicative.
 
BTW, if you extend that graph the formations that made the ANV stay pretty constant in strength until the autumn of '64 with dips after battles followed by a slow recovery (casualties returning). Antietam is special though.....
 
"Add back" is a reasonable technique.

Saying that no appreciable number of the wounded in the Maryland/Antietam/First Invasion/whatever campaign returned between September 2nd and October 10th is a bit far fetched.

If Thorp had used the September 22 figures and "added back" the losses of the 16th-20th (ideally 17th-20th, but we have little enough fighting on the 16th to not mar the overall accuracy by enough to change the odds) to determine a figure for the 17th that wouldn't be an issue.

But implying that none of the living from the losses after the beginning of September that are "added back" are included in the figure for the 10th (meaning that there's no double counting) is asking a lot of the reader's confidence in the writer's study of the subject. If someone wants to challenge the conventional wisdom then I want to see something better than the standard account, not just different. Taken on its own, Thorp's is "just different". And if we're going to use Harsh to verify, why should we use Thorp at all? What does he add that is more accurate or complete than Harsh?

And are you avoiding the question of where your definition of the term "present for duty" as "fit to march" is from, or am I just having a particularly bad case of the blinders as to where you answered?
 
Last edited:
Saying that no appreciable number of the wounded in the Maryland/Antietam/First Invasion/whatever campaign returned between September 2nd and October 10th is a bit far fetched.

Harsh notes (Taken at the Flood pg 123) Lee ordered for any convalescents from the Seven Days to be retained in Richmond rather than sent in dribs and drabs. Records begin in Sept '62, and aggregate strengths of the Richmond hospitals were:

Gen. Hosp. 1: 500
"" 2: 184
"" 3: 98
"" 4: 252
"" 5: 386
"" 6: 240
"" 7: 157
"" 8: 586
Clopton: 121
Gen. Hosp. 9: 888
"" 10: 219
"" 11: 268
"" 12: 324
"" 13: 251
"" 14: 410
"" 15: 331
"" 16: 392
"" 17: 209
"" 18: 312
"" 19: 277
"" 20: 351
"" 21: 466
"" 22: 162
"" 23: 140
"" 24: 244
"" 25: 130
"" 26: 86
"" 27: 31
Camp Winder: 6,021
Howard's Grove: 519
Camp Lee: 510
Chimborozo: 5,030
Louisiana Hosp: 298
South Carolina Hosp: 750
2nd Alabama Hosp: 261
3rd Alabama Hosp: 261
Robertson: 66
Henningsen: 97
Samaritan: 32
Bellevue: 46
St. Francis de Sales: 130
Medical College: 96
in Private qrs: 53

(exc. Federal prisoners in Military Prison Hosp.: 1,743 - almost all paroled that month)

Total aggregate sick in Richmond Sept '62: 22,165

Some convalescents at Leesburg (about a thousand) who dropped out around the Potomac were sent to the Winchester depot, and the stragglers depot at Winchester may have topped 6,000 at one point, but these were men who crossed the Potomac. No returnees from either Richmond nor Leesburg arrived before the 30th September state, so you're still in the 72,000 ball park. All these numbers are so close they're within what I'd call experimental error.
 
Some convalescents at Leesburg (about a thousand) who dropped out around the Potomac were sent to the Winchester depot, and the stragglers depot at Winchester may have topped 6,000 at one point, but these were men who crossed the Potomac.
Where did you come up with these numbers? Because I would put both numbers at least double.
 
What does the order for convalescents from the Seven Days to be retained at Richmond have to do with with the men wounded in the Maryland campaign?
 
Where did you come up with these numbers? Because I would put both numbers at least double.

At Winchester? They're the number John R. Jones reported there.

Expired Image Removed
Expired Image Removed

The point remains - all those men west of the Potomac had to have been with the army at Frederick for observation.
 
What does the order for convalescents from the Seven Days to be retained at Richmond have to do with with the men wounded in the Maryland campaign?

Quite a lot, because as I'm sure you know there is a constant in and outflow of troops due to sickness and wounds. When Lee crossed the Potomac the normal flow stopped. The troops at Winchester were those who dropped out of the army in Maryland.
 
You are misrepresenting what Jones reported.

Indeed, I was understating the case above.

Huh?[/QUOTE]

The crux of the argument that has developed is the reliability of all the observations of rebel strength in Maryland.

We have several calculated strengths from returns, all 75,000 =/- 5,000. We also have a number of observations of marching columns that are in similar ranges.

We're delving into the 152 year old issue of how Lee's army went from ca. 75,000 to barely 54,000 (deducting formations not at Antietam) that fought there. Livermore gives a slightly higher PFD of 55,221 at Antietam via add-back methodology (https://archive.org/stream/numberslosses00liverich#page/92/mode/2up ), which is being questioned.
 
Quite a lot, because as I'm sure you know there is a constant in and outflow of troops due to sickness and wounds. When Lee crossed the Potomac the normal flow stopped. The troops at Winchester were those who dropped out of the army in Maryland.

I presume that you're saying that all the troops who dropped out of the army in Maryland were at Winchester based on something you've already quoted, but how does the normal inflow or outflow of troops have anything to do with any troops wounded in Maryland and already in the ranks again by October 10, and thus not needing to be "added back"?
 
Mr. Thorp:

Elennsar, I have addressed your comments line-by-line below:

ELENNSAR:
Your first link (in your post above) asserting that we can determine Lee's starting strength by adding the campaign's causalities to the October return risks double counting anyone noted as a causality at some point during the campaign but who had returned to the ranks as of the October 10, or any men who had not been present at the start of the campaign returning to the ranks in time for the October 10 count being accidentally included as part of the campaign start count.

THORP:
I certainly hope you will read my first link because in it I have already addressed the potential issue of double counting.
http://www.washingtonpost.com/wp-srv/special/artsandliving/civilwar/mcclellan-graphic/?tab=2

ELENNSAR:
This is less troubling than reporting Lee's "infantry PFD" based on a return that includes two artillery battalions as well as divisional artillery (certainly it is not listed as missing, and is usually given when divisional totals are), but it is not any better than anyone else's guesstimate.


THORP:
"Infantry" present for duty should be "troops" present for duty.

The statement about Lee's Sept. 22 return in my article should read [changes in caps]:
Lee filed his first return five days after the battle, noting the count is "very imperfect" and does not include cavalry or RESERVE artillery. It states that on Sept. 22, he had at least 36,418 TROOPS present for duty.

I am unclear as to why you are so deeply troubled by this minor error. The fact that I wrote INFANTRY instead of TROOPS does not change my stated total for Lee's army of "about 41,000" on Sept. 22. Is it correct for you to encourage others to discount everything else I have written because of it?

ELENNSAR:
And your use of eyewitness head counts to justify claims of Lee's numbers does not address that even were those headcounts 100% accurate that they would be counting the total men marching by - a poor basis for determining Lee's present for duty as distinct from his aggregate present, and yet used as if these totals were both certainly reliable and certainly of present for duty only.

THORP:
If historians cannot consider eye witness accounts, first hand scout reports, Confederate tri-monthly reports and sworn statements made immediately afterwards by Union officers, then what are we supposed to use? The accounts I published are copied verbatim from these sources. You will note that most first-hand accounts are larger than the present for duty numbers that I used. Does that not reinforce the notion that these witnesses were seeing aggregate present? Where are your better sources to contradict mine? And where in my text do I claim that my totals are "certainly reliable and certainly of present for duty only?" My actual words were, "If more than 75,000 men were with Lee during the Antietam campaign, sightings of the different parts of Lee's army should generally match the estimated number of troops from the chart above. In the map below, two-dozen accounts, taken from newspapers, civilians, intelligence reports and the sworn statements of officers, are located on the march routes. The correlation is surprisingly close."
Expired Image Removed

It should be noted that additional eye witness accounts which I did not publish from Sept. 2 to Sept. 9, also correlate with Lee's estimated troop strengths. What is your superior method to calculate the size of Lee's army during that time period?

ELENNSAR:
I have no doubt about your interest in documenting Lee's numbers, but the soundness of your writing I find doubtful based on that being offered as supporting your conclusion.

THORP:
Again, what is your better method? Should we just not question Taylor's and Sears' numbers?

ELENNSAR:
As for the issue of Lee having lost more men than he is generally acknowledged as losing, that is not an area I have investigated myself to any extent, and given that it's a pretty recognizable fact that both sides undercounted their own losses it would not surprise me to see that happen here as well.

THORP:
So is it incorrect to add back in the extra losses to Lee's army? They had to have been present in Maryland during the time period from Sept. 13 to Sept. 18, did they not? Regarding Union troop losses during the Antietam Campaign, what evidence do you have that specifically shows the Union undercounted their losses anywhere close to the extent of the Confederates? A conservative estimate shows the rebels had at least 3,500 additional casualties and that is only including killed and prisoners of war. I do not even attempt to consider additional wounded, of which there certainly many .

ELENNSAR:
But it does not address the issue of the strength of his army being too much for McClellan's own forces, which is my main concern in regards to McClellan's actions, and thus is not one I find particularly convincing as relates to the defense of McClellan's actions.

THORP:
On Sept. 7 or there about, McClellan only had about 60,000 men. (I don't have the exact number at hand, but I can produce it.) If Lee had 75,000 present for duty in the Frederick area, then McClellan's caution at that time is more than justifiable. (Not withstanding Halleck's bickering that McClellan was moving too fast!) For the record, I do not advocate that McClellan could not have beaten Lee more soundly. However, judging by how harshly Lee handled Burnside, Hooker, and Grant in later conflicts, when the odds were much better for the Union, I think it is unrealistic to believe that Lee could easily have been destroyed at Antietam. What I am tired of is arm-chair historians constantly writing that throughout the campaign, McClellan outnumbered Lee 2 to 1 when Lee's own reported numbers do not support that conclusion.

ELENNSAR:
Frankly, I feel that neither this or Sears are convincing, and look to find better information in other sources when they're available.

THORP:
What new sources do you expect to become available? Again, what numbers are we supposed to use in the meantime and why?

This was all written in haste, but I would like to address your post of Aug. 29, 2014 next.
 
If you want a specific answer to a specific line I don't answer here I'll provide it, but this is what I have to say on my chief doubts:

THORP:
"Infantry" present for duty should be "troops" present for duty.

The statement about Lee's Sept. 22 return in my article should read [changes in caps]:
Lee filed his first return five days after the battle, noting the count is "very imperfect" and does not include cavalry or RESERVE artillery. It states that on Sept. 22, he had at least 36,418 TROOPS present for duty.

I am unclear as to why you are so deeply troubled by this minor error. The fact that I wrote INFANTRY instead of TROOPS does not change my stated total for Lee's army of "about 41,000" on Sept. 22. Is it correct for you to encourage others to discount everything else I have written because of it?

Let me illustrate with a later return for comparison's sake.

Lee's "grand total" on May 20 1863 (with one regiment and one company of cavalry not reporting, and one battalion of artillery and two other batteries not reporting) is 67,562 men strong.

Taking out the cavalry and the reserve artillery, that leaves us with 59,839 men.

An average infantry division - based on corps totals - is about 6900 men.

The difference between 59,839 and the total infantry (as artillery is listed separately) is almost two thirds of an average ANV division at this point.

Now, I don't know how many artillerymen are counted within the September 22 return. But if it's comparable like this, it makes the difference between - using your count for the troops not on the return - 36,418 infantry and an army total of "about 41,000" and an army total of a little less than thirty nine thousand - using the same figure for the not on the return.

Do you think two or three brigades of infantry make that little difference to the army's fighting power?

THORP:
If historians cannot consider eye witness accounts, first hand scout reports, Confederate tri-monthly reports and sworn statements made immediately afterwards by Union officers, then what are we supposed to use?
The accounts I published are copied verbatim from these sources. You will note that most first-hand accounts are larger than the present for duty numbers that I used. Does that not reinforce the notion that these witnesses were seeing aggregate present? Where are your better sources to contradict mine? And where in my text do I claim that my totals are "certainly reliable and certainly of present for duty only?"

Your statement that "sightings of the different parts of Lee's army should generally match the estimated number of troops from the chart above." refers to the chart where you attempt to calculate Lee's present for duty.

Trouble is, if Steiner's (for example) 69,000 is aggregate present, it doesn't really have a close relationship to 67,000 (not counting Walker and Stuart for comparison to Steiner's figure) present for duty at all. There's potentially not just a couple of brigades but a full division being added in effect here.

And if you don't consider them reliable, then why are they even included?
 
If you want a specific answer to a specific line I don't answer here I'll provide it, but this is what I have to say on my chief doubts:

So your objection is that the best available data is noisy? Let me put my working hat on, in fact the purple "Harry Potter" hood on my Ph.D. gown - welcome to real world data!

These numbers are all within statistical noise of each other. There is no significant difference.
 
67th Tigers: No.

My objection is that this is not an improvement on the known lowball estimates. It's just erring in the other direction and the eyewitness reports are even less useful at establishing present for duty than the reports giving effectives.

Using Thorp's "October 10 strength + campaign causalities returned" method Lee had around 91,700 men (aggregate present) at campaign start - in the interests of comparing apples to apples, not to oranges.

A week later, we have estimates from "over 100,000 strong" (on the 9th) to 60,000 (or potentially less per Miles, but it's not clear what portion of Lee's army is supposed to be counted here).

We still don't have actual reports on present for duty in this period.

Statistical noise can be ignored. Statistics that propose Lee's present for duty is a given number based on our available information on how many men were in the aggregate present column need to be looked at very carefully, not "Steiner's 69,000 at most marching by is close to 67,000 PFD, therefore the former bears out the latter."
 
To fully understand McClellan you need to study his service during the Mexican War. In Mexico City after the fighting was over McClellan was reported to have said that he was glad that no one could say "Poor Mac!" in reference that he had been killed in action. I believe that he lost whatever "killer" instinct that he had during the Mexican War.
 
To fully understand McClellan you need to study his service during the Mexican War. In Mexico City after the fighting was over McClellan was reported to have said that he was glad that no one could say "Poor Mac!" in reference that he had been killed in action. I believe that he lost whatever "killer" instinct that he had during the Mexican War.

I'm not sure that's entirely fair, or even at all fair.

I don't like McClellan, but I'm leery of casting aspersions on anyone's physical courage. McClellan's failures would not have been turned into successes by him eating some lead.
 

Learn About Us
About CivilWarTalk
Contact the Webmaster
Meet the Staff
Link to CivilWarTalk
Join Our Community
Register
Browse Forums
View Today's Discussions
Search the Forum
Get Help
FAQ
Student Guide
Forum Rules & Etiquette
Copyright / DMCA

     Contact Us CivilwarTalk on Facebook CivilWarTalk on YouTube CivilWarTalk on Twitter RSS Feed

Bringing the American Civil War and More to Life.
© 1999 - , CIVILWARTALK, LLC - Site Version 10.0

SlaveryTalk.com - SecessionTalk.com - CivilWarTalk.com - ReconstructionTalk.com
Back
Top