Confederate Strength 1862

My family on my fathers side has a similar tale. We can trace their history accurately from 1840 onwards, but that's only because some letters survive, but there's an amusing problem with them. Dutch immigrants (or Dutch descended) from the United States to Canada, some British civil servant seems to have heard their name, and written down permissions for them under a completely inaccurate one. The pre-existing "Van" got dropped from our names, and it was misspelled to make many people believe we were actually Jewish.

It makes for an... interesting attempt at research.



Part of the reason I'm curious of Confederate strengths is that, while it appears the Confederate army was large, I don't quite understand the impetus for the sudden roll of conscription acts that came in early spring 1862. If there was a sudden fear half the army was going to vanish it made sense, but if there was still going to be a vary large army left behind I was scratching my head and the new draconian conscription orders. Knowing how large the forces available to the Confederacy before all the big battles is quite helpful to me since it can give a scale of why more men were needed, and how many may have been lost in the fighting of 1862.
The CSA conscription story is extremely complex. Resistance to conscription in the Appalachian Mountains was fierce. Punitive expeditions were mounted by CSA conscription officials all up & down the mountains.

After the first blush voluntary enlistment in CSA regiments plummeted. Once an area of TN e.g. was lost to CSA control, local men went home. Few, if any, volunteers or draftees joined up to replace the steady hemorrhage of deserters.

The hard, cold reality is that without conscription the CSA army would have shrunk to a fraction. With mass desertion as time went on, even. Conscription could not fill the ranks.
 
I think the problem with continually restating that the numbers don't mean anything, or that the numbers don't translate to combat capability, or that the numbers are a starting point and nothing more, is that it's not really analysis - it's the abrogation of analysis.

It would certainly be valid to try and work out the actual state of the army based on reasonable assumptions, but one I like to point out is that Lee's army was considered (by Lee) to have insufficient officers to exercise proper control of the men in July 1862. This means that his army's "Present for Duty" in the category of "Men" is an under count relative to the Union definitions of the same, because Lee is reporting fewer men per officer than McClellan is at the same time as arguing that he has too many men per officer.
 
Whether it was conscription in the South or draft in the North most men were willing to 'volunteer' rather than wait to be forced into service.

*

From a report of February 1865. 82,000 represents less than 10% of the total enlistments of the Confederate army.

conscription.jpg

Official Records, Series 4, Volume 3, p.1101
 
I think the problem with continually restating that the numbers don't mean anything, or that the numbers don't translate to combat capability, or that the numbers are a starting point and nothing more, is that it's not really analysis - it's the abrogation of analysis.

It would certainly be valid to try and work out the actual state of the army based on reasonable assumptions, but one I like to point out is that Lee's army was considered (by Lee) to have insufficient officers to exercise proper control of the men in July 1862. This means that his army's "Present for Duty" in the category of "Men" is an under count relative to the Union definitions of the same, because Lee is reporting fewer men per officer than McClellan is at the same time as arguing that he has too many men per officer.
I would appreciate a citation that explicitly says that numbers don’t matter… of course there isn’t one. Instead of the habitual straw man, why not see my posts for what they really are? INTERPRETING the numbers is both the challenge & the way the raw numbers actually begin to mean something.
 
Last edited:
I would appreciate a citation that explicitly says that numbers don’t matter… of course there isn’t one. Instead of the habitual straw man, why note see my posts for what they really are? INTERPRETING the numbers is both the challenge & the way the raw numbers actually begin to mean something.


Well, there's a few which get close to that in language. I'll admit you didn't say the numbers didn't matter, but:
The lesson here is that nice neat numbers on a graph are not necessarily much use in determining the strength of an army.
What is the old saying? Numbers don’t lie, but people who compile them do. I have no idea what the troop strength numbers in 1862 are to be used for.
Just because a CSA regiment was listed as having X number present does not mean that the regiment had any combat capability.
Once again, a grid chart of numbers does not translate into combat capability.

Which seems to suggest that you consider the numbers to not actually mean very much and indeed you say you've no idea what the numbers are to be used for.

I will however concede that you appear merely to consider the numbers to be "not necessarily much use" and to "not translate" into combat capability.
 
Last edited by a moderator:
Present and fit for duty probably around 115-50 thousand. Another 100 thousand, awol , stragglers, sick or simply unfit for 19th century warfare hard to really know because of the unreliable returns.
I will say it again the army needed to be culled and rebuilt with a focus on quality not quantity as the confederates could never equal the union in numbers.
 
Present and fit for duty probably around 115-50 thousand. Another 100 thousand, awol , stragglers, sick or simply unfit for 19th century warfare hard to really know because of the unreliable returns.
I will say it again the army needed to be culled and rebuilt with a focus on quality not quantity as the confederates could never equal the union in numbers.
Those numbers seem terribly low - it implies an average Confederate regimental strength of less than 400 fit men per regiment before any of the fighting. I'd have put it closer to 700 fit men per regiment, which is the same sort of strength the Union had before any of the fighting.
 
Well, there's a few which get close to that in language. I'll admit you didn't say the numbers didn't matter, but you did say:






Which seems to suggest that you consider the numbers to not actually mean very much and indeed you say you've no idea what the numbers are to be used for.

I will however concede that you appear merely to consider the numbers to be "not necessarily much use" and to "not translate" into combat capability.
As usual “…seems to suggest…” rather than what was actually said. What I explicitly stated& gave unambiguous example of is that CSA troop numbers require very careful examination.

Nobody who is at all familiar with Wheeler’s returns takes a one of them at face value, e.g. The extreme difference between what Wheeler reported & the numbers he actually had drives historians to distraction. Drilling down to actual numbers is, I am assured, vexing in the extreme.

I don’t seem anything. I do exactly what historians do & analyze the numbers. Without that, numbers can seem to mean anything. Which we all know is the dictionary definition of a false premise.
 
Out of curiosity, would you be able to provide an example of Wheeler's returns? I've not actually seen them; should be interesting to see what he was reporting, versus what he actually had.
 
Out of curiosity, would you be able to provide an example of Wheeler's returns? I've not actually seen them; should be interesting to see what he was reporting, versus what he actually had.
Connelly, ‘Autumn of Glory’ I am not at home, but it won’t take you very long to look up Wheeler in the back & turn to Connelly’s detailed analysis & footnotes. As usual, Connelly was very thorough. I made copies of AG reports, which are especially damning. My copy has blue post it notes marking the pages & some very tattered hard copies from the OR enter leaved. Definitely a relic of another age of the world.

I haven’t studied Wheeler’s operations in detail for quite a while, but everything I ever read took note of both his inaccurate returns & magical thinking style reports.

Wheeler lived a long productive life & like any good Southern storyteller, he did not let the facts get in the way of his tales
 
So the example I found offhand was:

1622236121807.png



Which cites:

1622236145952.png


Which directs me to:

1622236262423.png


And it seems as though here the evidence used by Connelly about Wheeler's lack of strength relative to his theoretical strength is to look at his PFD (strictly his men PFD, at 6,594 for "sixty-six hundred") versus his Aggregate Present and Absent in his returns. So it appears as though this isn't a place where the Confederate returns are filled out incorrectly; instead Connelly's analysis is predicated fundamentally on the accuracy of the Confederate returns for Wheeler.


Later, on page 384, Connelly addresses Wheeler's available cavalry again:

1622236716096.png

For the second time, Connelly's source for this analysis is the actual OR rolls. They largely line up with the numbers he gives:

1622236873509.png

However there is one important caveat here. Connelly mentions the 2,419 number (given as Wheeler's cavalry effectives), the 8,062 number (PFD officers plus men) and the 18,785 number on the rolls (Wheeler's aggregate present and absent), but he's failed to notice (or at least note) that the total in the Effective Total Present column includes a blank for Martin's division. The comparison should thus exclude Martin's division and say that the units which had 2,419 effectives represented 12,299 men on the rolls and 6,136 men PFD. We don't know how strong Martin's division was in effectives.

I should however stress that Connelly's case here in both cases is based directly off the Confederate official tabulations of strength, and he's comparing Effectives with Aggregate Present and Absent on the same table - he is trusting the reports in both of these cases to be an accurate assessment of the numbers they say they are.

The use of AP&A is also something I'd never consider doing in the first place for any kind of evaluation of field strength, no matter the unit and no matter the time in the war - it counts men who aren't there, and this is just as true for the Union.


Now, I haven't been able to read the book through exhaustively (not all pages are available online) and it's quite possible that at a later point Connelly is citing an example of Wheeler's strength being different to what is in the actual reports of Confederate strength submitted by Wheeler or Wheeler's superiors in the ORs. But in the two cases I could find Connelly's assessment of Wheeler's strength was based on taking the OR values as read.

Am I missing something?
 
So the example I found offhand was:

View attachment 402300


Which cites:

View attachment 402301

Which directs me to:

View attachment 402302

And it seems as though here the evidence used by Connelly about Wheeler's lack of strength relative to his theoretical strength is to look at his PFD (strictly his men PFD, at 6,594 for "sixty-six hundred") versus his Aggregate Present and Absent in his returns. So it appears as though this isn't a place where the Confederate returns are filled out incorrectly; instead Connelly's analysis is predicated fundamentally on the accuracy of the Confederate returns for Wheeler.


Later, on page 384, Connelly addresses Wheeler's available cavalry again:

View attachment 402304
For the second time, Connelly's source for this analysis is the actual OR rolls. They largely line up with the numbers he gives:

View attachment 402305
However there is one important caveat here. Connelly mentions the 2,419 number (given as Wheeler's cavalry effectives), the 8,062 number (PFD officers plus men) and the 18,785 number on the rolls (Wheeler's aggregate present and absent), but he's failed to notice (or at least note) that the total in the Effective Total Present column includes a blank for Martin's division. The comparison should thus exclude Martin's division and say that the units which had 2,419 effectives represented 12,299 men on the rolls and 6,136 men PFD. We don't know how strong Martin's division was in effectives.

I should however stress that Connelly's case here in both cases is based directly off the Confederate official tabulations of strength, and he's comparing Effectives with Aggregate Present and Absent on the same table - he is trusting the reports in both of these cases to be an accurate assessment of the numbers they say they are.

The use of AP&A is also something I'd never consider doing in the first place for any kind of evaluation of field strength, no matter the unit and no matter the time in the war - it counts men who aren't there, and this is just as true for the Union.


Now, I haven't been able to read the book through exhaustively (not all pages are available online) and it's quite possible that at a later point Connelly is citing an example of Wheeler's strength being different to what is in the actual reports of Confederate strength submitted by Wheeler or Wheeler's superiors in the ORs. But in the two cases I could find Connelly's assessment of Wheeler's strength was based on taking the OR values as read.

Am I missing something?
If you have not read ‘Army of the Heartland’ & ‘Autumn of Glory’ you have missed some of the seminal scholarship of CW history. There is a very good reason why those books have never gone out of print while so many others only exist in the five books for a dollar bin.

Connelly wrote at a time when the CW was still a living memory. When first published, the two volume bio of the Army of Tennessee set the Lost Causers’ hair on fire. He gored a whole herd of sacred cows.

I am sincere when I say that I am very surprised that you haven’t studied Connelly. Scholarship of the Western Theater springs directly from his seminal research. Connelly is footnoted everywhere. It is not too much to say that if you haven’t read Connelly, you can’t understand the Army of Tennessee & the War in the West.

Connelly wrote during the age when the only way to access documentation was the mark one eyeball. He had some very dedicated research assistants with deep knowledge of the archives. I personally have benefited greatly from following his trail via the internet.

I am not being rude when I say the a good many of your seems & guesses wouldn’t exist if you had studied Connelly.
 
If you have not read ‘Army of the Heartland’ & ‘Autumn of Glory’ you have missed some of the seminal scholarship of CW history. There is a very good reason why those books have never gone out of print while so many others only exist in the five books for a dollar bin.

Connelly wrote at a time when the CW was still a living memory. When first published, the two volume bio of the Army of Tennessee set the Lost Causers’ hair on fire. He gored a whole herd of sacred cows.

I am sincere when I say that I am very surprised that you haven’t studied Connelly. Scholarship of the Western Theater springs directly from his seminal research. Connelly is footnoted everywhere. It is not too much to say that if you haven’t read Connelly, you can’t understand the Army of Tennessee & the War in the West.

Connelly wrote during the age when the only way to access documentation was the mark one eyeball. He had some very dedicated research assistants with deep knowledge of the archives. I personally have benefited greatly from following his trail via the internet.

I am not being rude when I say the a good many of your seems & guesses wouldn’t exist if you had studied Connelly.
I'm going to ask again - I was after an example of Wheeler's returns, and how they didn't reflect what he actually had. Looking at the Connelly book you pointed me at, I have found two examples of where Connelly is addressing the question of how many troops Wheeler had by taking Wheeler's returns at face value.

Am I missing something here? That is, a reading of the book you pointed me at seems to suggest that Wheeler's returns are accurate, in that Connelly does not question them or rather relies on them to be accurate; is this correct, or is this something I am in some way missing?

What you said was:
Nobody who is at all familiar with Wheeler’s returns takes a one of them at face value, e.g. The extreme difference between what Wheeler reported & the numbers he actually had drives historians to distraction. Drilling down to actual numbers is, I am assured, vexing in the extreme.
But I'm now finding Connelly using Wheeler's returns at face value. I assume I'm missing something.
 
What makes this thread a delight is that people played with numbers back then just like today. I do appreciate Bragg saying he didn't know how many soldiers he had or where they were to no less than the President, that kind of admission is usually a career ender.
 
What makes this thread a delight is that people played with numbers back then just like today. I do appreciate Bragg saying he didn't know how many soldiers he had or where they were to no less than the President, that kind of admission is usually a career ender.
I agree, I have never discussed this with anyone who is not baffled by Davis’ action. There is no record of their conversation. Bragg was notorious for pissing people off ‘ Davis was notoriously touchy. How does walking into the room after a debacle in Kentucky & admitting that you haven’t a clue where your army is result in a resounding affirmation? It passes all understanding. Of course, Davis had been A. S. Johnston’s enabler… go figure.
 
I'm going to ask again - I was after an example of Wheeler's returns, and how they didn't reflect what he actually had. Looking at the Connelly book you pointed me at, I have found two examples of where Connelly is addressing the question of how many troops Wheeler had by taking Wheeler's returns at face value.

Am I missing something here? That is, a reading of the book you pointed me at seems to suggest that Wheeler's returns are accurate, in that Connelly does not question them or rather relies on them to be accurate; is this correct, or is this something I am in some way missing?

What you said was:

But I'm now finding Connelly using Wheeler's returns at face value. I assume I'm missing something.
Read the books. Wheeler’s screw ups extend all the way back to 1861. Believe you me, taking on Connelly without reading every page of the books is a mistake. You will be talking to a host of people who know volumes that you do not. I mean this in the friendliest way possible. The usual straw men & speculation will be met with laughter.
 
Read the books. Wheeler’s screw ups extend all the way back to 1861. Believe you me, taking on Connelly without reading every page of the books is a mistake. You will be talking to a host of people who know volumes that you do not. I mean this in the friendliest way possible. The usual straw men & speculation will be met with laughter.
I want to be absolutely clear about this, because I'm not taking on Connelly* - what I'm doing is questioning whether the source you pointed me at is one I'm reading correctly.

* except in the most trivial sense, such as pointing out that columns on a table he references don't seem to have been added in a consistent way, because I can see how the numbers add up.


What I'm doing is asking how what I have excerpted (mentions of Wheeler in the context of Wheeler's official returns, where the official returns are used as data) aligns with your statement that "nobody who is at all familiar with Wheeler takes Wheeler's numbers at face value".

Connelly says in so many words that on May 1 1864 Wheeler could only muster 2,419 effectives, and his citation link is to the ORs; reading the ORs shoes that the 2,419 number is one that is literally in the OR table from Wheeler's returns.


1622247990365.png

1622247998390.png


What this looks like to me is that Connelly is taking that number from Wheeler's returns. Is that not exactly what you said nobody would do without checking? What have I missed?
 
I want to be absolutely clear about this, because I'm not taking on Connelly* - what I'm doing is questioning whether the source you pointed me at is one I'm reading correctly.

* except in the most trivial sense, such as pointing out that columns on a table he references don't seem to have been added in a consistent way, because I can see how the numbers add up.


What I'm doing is asking how what I have excerpted (mentions of Wheeler in the context of Wheeler's official returns, where the official returns are used as data) aligns with your statement that "nobody who is at all familiar with Wheeler takes Wheeler's numbers at face value".

Connelly says in so many words that on May 1 1864 Wheeler could only muster 2,419 effectives, and his citation link is to the ORs; reading the ORs shoes that the 2,419 number is one that is literally in the OR table from Wheeler's returns.


View attachment 402314
View attachment 402315

What this looks like to me is that Connelly is taking that number from Wheeler's returns. Is that not exactly what you said nobody would do without checking? What have I missed?
It is pointless to play word games. I am explicitly saying that until you have read Connelly cover to cover the numbers, wherever Connelly found them are meaningless. Forensic accounting has its place, but in isolation it is “ the high road to folly.”

This thread is about CSA troop strength in 1862. I am not an archivist, so your are wasting time asking me about Connelly’s methodology. I do know that he has a reputation as a scholar, so am confident his sources are the best that could be had at the time. We have access to data he did not have. Dissertations have been written on that topic.

Bottom line, to continue the accounting analogy, is that until you have studied Connelly, the numbers are not going to mean very much.
 
It is pointless to play word games. I am explicitly saying that until you have read Connelly cover to cover the numbers, wherever Connelly found them are meaningless. Forensic accounting has its place, but in isolation it is “ the high road to folly.”

This thread is about CSA troop strength in 1862. I am not an archivist, so your are wasting time asking me about Connelly’s methodology. I do know that he has a reputation as a scholar, so am confident his sources are the best that could be had at the time. We have access to data he did not have. Dissertations have been written on that topic.

Bottom line, to continue the accounting analogy, is that until you have studied Connelly, the numbers are not going to mean very much.
What I was interested in was seeing an example of Wheeler's reports and how they were out of step with reality - as you stated was the case here:

Nobody who is at all familiar with Wheeler’s returns takes a one of them at face value, e.g. The extreme difference between what Wheeler reported & the numbers he actually had drives historians to distraction.

And here:
The king of confusing returns was Joe Wheeler. At one point, IG found that the number of troopers mounted & ready for duty was 10% of Wheeler’s reported strength.

And here:
On the last day of 1862, Bragg lost about1/3rd of his army on the first day of Stones River. It would be weeks before an accurate count of the strength of the A of TN. Folks who study Wheeler report that as usual, his returns are wildly inaccurate.


And when I asked, you said:

Connelly, ‘Autumn of Glory’ I am not at home, but it won’t take you very long to look up Wheeler in the back & turn to Connelly’s detailed analysis & footnotes. As usual, Connelly was very thorough.

That I should read Autumn of Glory, look up Wheeler in the back and turn to Connelly's detailed analysis and footnotes. I am going through Autumn of Glory, looking up Wheeler and going to Connelly's detailed analysis and footnotes, and what I am finding is cases where Connelly is highlighting Wheeler's (reported) effective strength being much lower than his (reported) aggregate present and absent. I am also finding cases where Wheeler is supposed to have been claiming his cavalry was in fine condition (in letters that are part of the Bragg papers, which I can't access).

What I'm not finding, however, is an example of Wheeler's returns being wildly inaccurate. Instead, whenever I find discussion of Wheeler's returns, they are treated as accurate - here's another example:

1622288543189.png


The 2,392 number is from the same table I excerpted before - it's the 2,419 number minus the 27 artillery effectives. In this case Connelly does mention Martin's detached division, and the mention of Wheeler's estimate implies Martin's division was at about 1,900 effectives at the end of April 1864.


On page 386 there is a discussion of cavalry strength:

1622288874482.png

But the general thrust of this passage is that Johnston's returns show that he had more men than he claimed. And Connelly is explicit that of the two of these (Johnston's returns or Johnston's complaints about lacking cavalry) he believes Johnston's returns, which is why Connelly states that "Johnston and later historians exaggerated his cavalry shortages".

Later (after the 1864 raid) Connelly says that Wheeler claimed 2,000 men with him at Tuscumbia, giving a citation in the ORs, but I can't locate the 2,000 figure at his citation; it's possible that the 2,000 figure is in the other citation given (Dyer's "Life of Wheeler") but if the only case of a possibly mendacious figure isn't in the ORs then, well, it isn't in the ORs.



When I started going through this book as you advised, what I was expecting to find was something along the lines of:

"Wheeler claimed 5,000 effectives in his report of 20 November 1862, but reports from XYZ indicate that he had no more than 2,000 horses and so he could not really have had 5,000 effective cavalry on that date"

But that's not what I've found.

I assume I'm missing something, but I am doing exactly as you advised (to whit, reading through the book and looking at where Connelly examines Wheeler's reported numbers) and in every case I can find where Connelly addresses the question of Wheeler's numerical strength the OR strength reports are what is treated as true.
 
Last edited by a moderator:
Has it crossed your mind that the massive difference between the numbers is the point?

If the numbers Connelly used were not correct then the conclusions would be based on a false premise. That, of course, leads to false conclusions.

Hundreds of thousands of his readers know that Connelly’s research was meticulous. You have now joined that host. Welcome to the club.

You can now read the books secure in the knowledge that Connelly’s reputation as a first class scholar is deserved. After reading the books you will understand why Connelly used those citations, which is after all, the whole point of the exercise.
 
Back
Top