Union numbers, strength reporting, and losses in the ACW (long post)

1 view
Skip to first unread message

Dimitri Rotov

unread,
May 31, 2004, 1:35:34 PM5/31/04
to
Everyone has a litmus test or two for analyzing a browsed book that
does not sit too squarely in one's own area of expertise. Mine is,
"How is this author treating numbers?" If I see a lot of zeroes, the
book gets demerits. If I see zeroes with a footnote, I may soften up
for an impulse buy. If I see very few zeroes and explanatory for the
number, I give top marks and my interest in the non-numerical analysis
is raised.

Some authors who sling zeroes will justify themselves by saying that
it is inherently dishonest not to estimate; that a number without lots
of zeroes represents a pretense. I disagree. The number with very few
zeroes is not presented as a fact, but as the very best estimate
available. The number with lots of zeroes, always present in pop
history, represents the pig in the poke you are supposed to buy with
no background or supporting info.

I thought I would toss out some of my prejudices to the group for what
they are worth as entertainment. These are not positions I necessarily
intend to argue here, but they are biases that guide me in Civil War
analysis.

1) There are official monthly returns filed for Union and
(intermittently) Confederate strengths. These returns are not only the
basis for many history estimates but also for remarks made by the
commanders in letters and chit chat. They are, what we would call
today, "the party line" on strengths and they are based on elected,
state officers' self-serving head counts. The monthly payroll muster
returns represent carefully constructed lies to ensure that everyone
in the unit who could possibly claim pay gets pay; they are maintained
and submitted by the officer who was elected by the payee. In modern
terms, they represent the equivalent of unit readiness reporting - and
they are about as reliable.

(2) These monthly payroll returns are false on a number of points,
including commander level CYA, favoritism, illegal absentee leave
(leaves granted by subordinate commanders not authorized to grant
leave - a huge problem in the ACW) and straggling. The higher federal
commanders knew that at state/regiment level, the returns were being
spun in favor of absentees.

(3) On the peninsula, to use an example I am familiar with, the
*daily* (non-payroll, non-monthly) returns provided by Union units
(never, ever cited by historians but available in Washington) show
present for duty numbers at a rate of 60% or less than the amount
stated on the monthly (payroll) musters. You must let that sink in.
Put another way: The daily returns are completely and totally out of
synch with the monthly payroll returns on which aout battle-centric
historians calculate present for engagement. And calculate tactical
merit. It is from this reduced (daily) figure that any additional
desertion and straggling must be calculated. Let that sink in too. The
Union levels of desertion and straggling were horrendous, and I'm not
talking about desertion and straggling that occurred before the 60% of
payroll muster regiment wheels into line, I'm talking about desertion
and straggling from the reduced strength unit engaged in combat. If
you don't want to travel to DC to look at daily rolls, look at the
Surgeon General's multivolume report on the war of the Rebellion - the
surgeons were mandated never to use the monthly payroll figures so
beloved of pop historians - they were to calculate incidents of
disease, wounds, and death from the actual boots on the ground - the
daily muster. Take a look at their strength figures to get grounded in
some real numbers.

4) McClellan and Lincoln had an ongoing dialog about this disparity;
Grant hints at it in his memoirs (see postscript below). It was
something Mac monitored closely and Lincoln's quip about shoveling
fleas was in the context of their ongoing discussion. By the end of
the Battle of Antietam, Mac had Meade, commanding Hooker's Corps,
prepare a secret memo for Lincoln's visit telling how many men Hooker
had brought into battle versus what the returns (and today's
historians) said he brought to bear. Likewise, in a conversational
slip talking to some ladies at the White House, Lincoln indicated he
knew the real (shocking) Union strengths versus those shown on returns
and in newspapers and that there was an awful problem. I believe this
dialog may have continued after McClellan.

(5) Fox generally and Harsh for Maryland have shown how ludicrously
out of synch historians' estimates of Rebel strengths are when
matched against the force structure deployed with them.

(6) The Republican corps commanders on the Peninsula told Halleck
before the withdrawal that they had fought outnumbered throughout. Our
habit is to ridicule these combat veterans because we have seen the
monthly payroll figures - the ones they signed! - and we cannot
imagine higher Confederate strength figures than the numbers reported
at payroll muster (score one for the Lost Cause). The idea that
depleted Union forces fought comparable Rebel TO&E, the rebels having
the higher fill rate per unit, is way, way over the heads of today's
generation of historians.

(9) Such record as was left by corps commanders after Antietam
indicates they thought the fight was even (Hooker) or that the Union
was outnumbered (Sumner, Franklin). Re blue vs gray estimates,
historians consistently disparage the battlefield estimates made by
experienced commanders in favor of hypothetical strength figures
derived from short, simple, careless paper drills that produce lots of
zeros at the end of a figure.

I'll leave you with some figures compiled by the regimental surgeons
that will show how sad is the state of Civil War history.

From The Medical and Surgical History of the War of the Rebellion
(1861-1865)," GPO 1875 "Second Issue" prepared under the direction of
Surgeon General
Joseph K. Barnes, US Army, by JJ Woodward, Asst. Surgeon, US Army.


"This figure [mean strength] was invariably obtained by adding
together the mean strengths given in the individual reports
consolidated for the month. It is simply to be understood as
representing the number of men among whom the diseases and deaths ...
occurred. No other plan would have served as the basis for the
deduction of the correct ratios [healthy to ill] from the tables."
(page xxii)

Think about what you just read.

"During a part of the war the medical officer of each regiment or
detachment was ordered to obtain the mean strength of the command
represented in his report by adding together the strength present, as
obtained from the consolidated morning report of the command, for each
day of the month, and dividing by the number of days." (page xxiii)

So, let's go. Army and Department of the Potomac:


8/61 - 50,608
9/61 - 85,408
10/61 - 113,204
11/61 - 133,669
12/61 - 152,759
1/62 - 167,267
2/62 - 153,308
3/62 - 126,588
4/62 - 71,250
5/62 - 72,536
6/62 - 78,733
7/62 - 106,069
8/62 - 69,320
9/62 - 149,052 (amalgamated armies and departments - DR)
10/62 - 171,258 (amalgamated armies and departments, start of Second
Richmond Campaign - DR)

Understand that the number next to 4/62, for instance, presents a
picture of the manpower pool available on any given day in April 62
(71,250), from which you then take out the cooks, sentries, ill,
stragglers, messengers, etc.

I'll close with a hunch. If in the East or West you have signed pay
authorizations for over 100,000 troops but you have a roll call in the
morning that shows 71,250, for instance; and if 20% or more of those
straggle moving into battle, then yes, you may very well fight
outnumbered. How do you represent that publicly? Through an enemy
strength estimate that is a multiple of your paper figure.

This is why I think so many Eastern and Western generals overstated
enemy strengths - they were expressing in ratio form a truth tempered
to align with the phony strengths they themselves reported via the
flawed payroll muster figures.

In interviews with FM Wolseley after the war, in discussing Antietam
in particular, Lee points to the same problem on the Rebel side.

Dimitri
http://cwbn.blogspot.com

p.s. Since writing this, I've found some specifics alluded to
generally above. I'll append them rather than rewrite the post. They
are from Lonn's book, Desertion During the Civil War:

* In speaking to some ladies after Antietam, Lincoln told them that
McClellan lost 30,000 men to straggling within two hours of the start
of the battle.

* At least 70,000 AoP men were at home on illegal fuloughs issued by
company grade officers who lacked the authority to grant such leave.

* By 1864, Grant was saying that out of every five men enlisted he got
the services of one soldier. (paraphrased from Lonn, Desertion During
the Civil War)

Bruce Smith

unread,
May 31, 2004, 7:15:38 PM5/31/04
to

"Dimitri Rotov" <roto...@mailcity.com> wrote in message
news:8ecc5305.04053...@posting.google.com...

Not sure what your point here is, Dimitri? The Official Records give the
returns as recorded, over a multi-year research project under the auspices
of the Federal Government. If you have done research in the National
Archives which disputes this, that would be interesting. Please quote this
research. If you have done research in State Archives - which can also
conflict with the National Archives - please quote this. Your research
would be quite interesting. This sounds like "secondary source" research,
and not "primary source." Please corroborate.
Kind Regards.
Bruce Smith.


Dimitri Rotov

unread,
Jun 1, 2004, 5:27:00 PM6/1/04
to
"Bruce Smith" <hanfeld...@comcast.net> wrote in message news:<Ya6dndhdepg...@comcast.com>...
<snip>

> Not sure what your point here is, Dimitri?

To quote me:

> > I thought I would toss out some of my prejudices to the group for what they are worth as entertainment. <<

Bruce said:
> The Official Records give the
> returns as recorded, over a multi-year research project under the auspices
> of the Federal Government.

Monthly payroll returns, Bruce. Compiled by elected officers. Nothing
like morning reports.

Bruce:


> If you have done research in the National
> Archives which disputes this, that would be interesting. Please quote this
> research.

To quote myself again:

> > From The Medical and Surgical History of the War of the Rebellion
(1861-1865)," GPO 1875 "Second Issue" prepared under the direction of
Surgeon General Joseph K. Barnes, US Army, by JJ Woodward, Asst.
Surgeon, US Army. <<

> If you have done research in State Archives - which can also


> conflict with the National Archives - please quote this. Your research
> would be quite interesting.

It would be interesting. I should have started when I was 20 because
that may be where the USV regiments' morning reports have ended up.
However, in the meantime, the regimental surgeons were good enought to
calculate means from daily returns that we can use in presenting
military history correctly. Remember that the figures in the Official
Records that you prize were rejected by the surgeons (one of the
points of my post).

> This sounds like "secondary source" research,
> and not "primary source." Please corroborate.

It's an interesting question whether the regimental surgeon's strength
numbers are secondary if they represent means taken from the daily
morning reports. If they are secondary, they are no more or less
secondary than the payroll muster figures given in the OR.

> Kind Regards.
> Bruce Smith.

Best to you, sir,

Dimitri
http://cwbn.blogspot.com

Howard G

unread,
Jun 2, 2004, 7:52:35 AM6/2/04
to
Dimitri Rotov wrote:

> Everyone has a litmus test or two for analyzing a browsed book that
> does not sit too squarely in one's own area of expertise. Mine is,
> "How is this author treating numbers?" If I see a lot of zeroes, the
> book gets demerits. If I see zeroes with a footnote, I may soften up
> for an impulse buy. If I see very few zeroes and explanatory for the
> number, I give top marks and my interest in the non-numerical analysis
> is raised.

To me, it's according to the type of book being read, the context,
according to whether the numbers are found within a statement, and then
conditionally, or are listed in tabular form.

Specific numbers such as "12,912 Union prisoners died at Andersonville" is
acceptable as is the round figure in "more than 33,000 men incarcerated in
the camp in August 1864" and "[n]early 13,000 of the 45,000 men who
entered the stockade died there." As a general statement, "there were
4,000,000 slaves at the beginning of the Civil War" would be acceptable,
but a tabular listing of slaves by states in 1860 containing rounded
numbers would be suspect as census specifics are available.


Bruce Smith

unread,
Jun 3, 2004, 5:53:58 PM6/3/04
to

"Dimitri Rotov" <dro...@gmail.com> wrote in message
news:d3e1218b.04060...@posting.google.com...

> "Bruce Smith" <hanfeld...@comcast.net> wrote in message
news:<Ya6dndhdepg...@comcast.com>...

<snip>

> > If you have done research in State Archives - which can also


> > conflict with the National Archives - please quote this. Your research
> > would be quite interesting.
>
> It would be interesting. I should have started when I was 20 because
> that may be where the USV regiments' morning reports have ended up.
> However, in the meantime, the regimental surgeons were good enought to
> calculate means from daily returns that we can use in presenting
> military history correctly. Remember that the figures in the Official
> Records that you prize were rejected by the surgeons (one of the
> points of my post).

I did extensive work on the 5th NH a number of years ago in the Concord, NH,
State Archives. I seem to recall the daily reports there. Be warned: most
State Archives in my experience are notoriously unorganized and
uncatalogued. I was handed five shoe-box-sized collections of uncollated,
unorganized documents and told to have fun. It is very tedious, but the
effort pays off, because one gains a depth of understanding unavailable to
the student of ACW who merely reads a textbook or a memoir.


> > This sounds like "secondary source" research,
> > and not "primary source." Please corroborate.
>
> It's an interesting question whether the regimental surgeon's strength
> numbers are secondary if they represent means taken from the daily
> morning reports. If they are secondary, they are no more or less
> secondary than the payroll muster figures given in the OR.

The "primary source" would be the document itself. A published collection
is a secondary source. However, the "Official Records" has come to attain
the status of "primary source" among most historiacal scholars, as long as
the original document can be proven to exist.

Kind Regards.
Bruce Smith.


Dimitri Rotov

unread,
Jun 5, 2004, 9:53:37 PM6/5/04
to
"Bruce Smith" <hanfeld...@comcast.net> wrote in message news:<ruidnYLLsIp...@comcast.com>...
<snip>

> The "primary source" would be the document itself. A published collection
> is a secondary source. However, the "Official Records" has come to attain
> the status of "primary source" among most historiacal scholars, as long as
> the original document can be proven to exist.
>
> Kind Regards.
> Bruce Smith.

Hi Bruce. Under those rules, which are long familiar to me, I duly and
publicly claim "primary source status" for the Surgeon General's OR.
Just don't ask me where the morning reports are filed!

What this has to do with our conversation is this: I claim that the
surgeon general's OR numbers (primary source) are more accurate than
the payroll muster figures we all see in the War Department OR (also
primary source). That's our disagreement, if it is a disagreement.

If, going through your New England regiment's records, you tallied
strength figures from the *morning reports*, it would be fascinating
to match those against the reported *payroll muster* figures reported
to HQ.

Best to you,

Dimitri
http://cwbn.blogspot.com

Reply all
Reply to author
Forward
0 new messages