Wednesday, 3 February 2016

US 1898 Mobilisation

A few facts for somewhere else. These are the raw numbers for the US mobilisation of 1898:

Table 1: Authorised Strength

Table 1 shows the authorised or establishment strength of the US Army. The regulars had a massive expansion (on 1st April 1898 it had only 28,183 officers and men) and never made authorised strength. The volunteers did make roughly authorised strength over a few months.

Table 2: Strength of US Army by last day of the month

Unfortunately the tables I extract this from don't divide the regular artillery into light artillery with guns (on wars outbreak 10 batteries ISTR, 2 each from 1st-5th US Arty), heavy artillery manning fortifications (6th and 7th US Arty) and those acting as infantry (majority of 1st-5th US Arty).

By the end of August 274,712 officers and men had enrolled, meaning the US basically met it's target.

The strength of the 7 army corps stood up (VIth Corps was never activated) was:

Table 3: Strength of Corps (Enlisted only)

Sunday, 24 January 2016

Strength of Canadian Militia ca. 1862


Over on Smith is once again arguing for the weakness of the Canadian militia citing Martin Petrie's book as apparently definitive. This is essentially a case of confirmation bias and a misreading of evidence.

The "Evidence"

It's worth noting that Petrie probably took his figures from the United Service Magazine, which was quoting a newspaper thus:

"The present Volunteer force of Canada consists of 10,615 infantry, 1,687 artillery, 1,615 cavalry and 202 engineers. There are also corps drilled and only waiting the recognition of the Government to be armed. This will swell the Volunteer force to 16,000 men. There are also in Canada 10,000 Militia. The whole military force of Canada will consist shortly of 26,000 troops."

Which itself is a corruption of this printed in the Illustrated London News on 13th September 1862:
"From an official return lying before us (say the Canadian) we find that the total volunteer force of Canada now in existence and recognised by government number 14,213.  Of these 10,615 are infantry, 1,687 are artillery, and 1,615 are cavalry with 202 are engineers. There are still in addition other corps drilled and only waiting the recognition of the government to be armed. These unrecognised companies would swell the aggregate volunteer force of the province to over 16,000 men; and, if to these are added the 10,000 militia provided by the new Act, we have a gross total of 26,000 men."

This is likely an accurate statement of the number of volunteer troops in the Province in Canada at the time. However, what is that militia number? The Canadian Government gives the strength of the militia at this time as around 470,000 (ref), so who are these "10,000 militia"?

To answer the question one must look to the 1862 Militia Bill, which lead to the fall of the incumbent Provincial Government in May. It proposed calling out for training over 100,000 militiamen for 12 days (more for artillery etc.) annually, but the bill was defeated on the basis that it would cost a million dollars a year and would require a tax increase. Emergency legislation empowered the government to pay for the call out 10,000 service militia (i.e. non-volunteers) per year for training, which is what the "10,000 militia" were.

On this basis Smith rejects the notion that if an emergency came (such as a threatened US invasion) the Canadian government could embody any additional troops beyond those that underwent paid training in the last year. If one were to apply this standard to the US of 1861 then only a single regiment would be available, the 7th New York State Militia, who paid for themselves to train with the fortress artillery in NY Harbor - indeed no northern state government had paid for a militia call out since 1856 (when NY paid to call out the militia for 6 days training).

Applying Smith's Argument to the USA.

Assume we don't know about the mobilisation of volunteers as actually happened.

The strength of the (Northern) US Volunteer Force in 1861 was 41,190 on paper, as distributed in this table from Todd (Mil. Affairs, 5(3), 152-170):

Thus the volunteer force of United States is roughly 1% of their male population of military age, which is oddly close to the ratio for Canadian active militia.

The government managed to call out 29,714 in these existing volunteer organisations, although this overstates the case. New York sent 10,824 militiamen into service (inc. 4 regiments for the war), but only about half of them were existing militiamen. In 1860, the state of NY inspected 11,149 offrs and men out of 19,189 enrolled (58.1%). If that ratio extended to the other states then about 24,000 men turned up for their once a year afternoon muster. As mentioned above, no state had provided the money for training (6 days in NY, similar in others) since 1856.

As well as the existing organisations, the militia formed new "volunteer" regiments by simply recruiting, and furnished the Federal government 46,709 volunteers, as per Todd:

Now, of the 76,423 30 and 90 day militia embodied, only 39% come from existing organisations, and about half the men in those were pre-crisis militiamen. If TS's logic is applied then this could not have happened!

So, Canada....

The experts continued to believe 100,000 men could be called out, and HMG provided over 100,000 Enfields etc., along with large quantities of artillery to equip them. Incidently, here one notes that TS quotes Bourne on the number of weapons in Canada in 1861, but these are "Imperial" weapons belonging to HMG, rather than Canada. Senior (ref) notes the Province of Canada acquired 50,000 surplus percussion muskets for the militia in the 1850's, and this is in addition to existing weapons - the Canadian government armed 46,222 militia infantry in 1837.

Given the fact that the population of the Province of Canada had more than doubled since 1837, one must conclude that the idea of embodying 100,000 militia in an emergency is not a flight-of-fancy, but in fact a repetition of 1837 mobilisation levels.

Wednesday, 23 December 2015

RN Steamships in Reserve December 1861

On my harddrive I found a list I compiled of the Steam Reserve at the time of the Trent Affair. These were the Royal Navy's reserve ships. Those in the 1st class are basically ready to go after having a crew assigned and loading gunpowder. The 2nd class are not stored or coaled, and thus require such. Those in the 3rd class require minor work before Commission (such as repairing HMS Black Prince's mast).

Ships requiring major repairs are paid off into Ordinary and are were not considered part of the Steam Reserve.

1st Class Reserve (Ready for immediate Commission)

1.    Duncan
2.    Princess Royal
3.    Meeanee

1.    Orlando
2.    Euryalus
3.    Severn

Corvettes, Sloops etc.
1.    Barrosa
2.    Chanticleer
3.    Petrel
4.    Rosario
5.    Stromboli
6.    Devastation
7.    Vigilant
8.    Victor
9.    Pandora
10.    Sparrow
11.    Lee

and 18 screw gun-boats

2nd Class Reserve (Require stores to Commission)

1.    Duke of Wellington
2.    Royal Sovereign
3.    Gibraltar
4.    Hood
5.    Rodney
6.    Royal William
7.    Bombay
8.    Goliath
9.    Lion

1.    Phoebe
2.    Sutlej
3.    Newcastle
4.    Galatea

Corvettes, Sloops etc.
1.    Zebra
2.    Rapid
3.    Coquette
4.    Cormorant
5.    Eclipse
6.    Lily
7.    Racehorse
8.    Serpent
9.    Star
10.    Dart
11.    Mullet
12.    Snipe
13.    Speedwell
14.    Vesuvius

and 18 screw gunboats

3rd class Reserve (Require repair before Commission)

Armoured Frigates and Batteries
1.    Black Prince
2.    Defence
3.    Resistance
4.    Erebus
5.    Glatton
6.    Thunder
7.    Trusty

1.    Prince of Wales
2.    Howe
3.    Royal Albert
4.    Victoria
5.    Windsor Castle
6.    St Jean d’Acre
7.    Anson
8.    Atlas
9.    Orion
10.    Renown
11.    Albion
12.    Frederick William
13.    Nelson
14.    Prince Regent
15.    Prince George
16.    Waterloo
17.    Brunswick
18.    Centurion
19.    Collingwood
20.    Cressy
21.    Irresistable

1.    Arethusa
2.    Aurora
3.    Bristol
4.    Chesapeake
5.    Constance
6.    Glasgow
7.    Leander
8.    Liverpool
9.    Octavio
10.    Undaunted
11.    Tribune
12.    Highflier
13.    Furious (Paddle)
14.    Leopard (Paddle)
15.    Magicienne  (Paddle)
16.    Retribution (Paddle)
17.    Penelope (Paddle)
18.    Valorous (Paddle)
19.    Eurotas (fitted as Mortar Frigate)
20.    Forth (fitted as Mortar Frigate)
21.    Seahorse (fitted as Mortar Frigate)
22.    Horatio (fitted as Mortar Frigate)

Sail Frigates

1.    Rattlesnake

Corvettes, Sloops, etc.

1.    Esk
2.    Pylades
3.    Alert
4.    Archer
5.    Cruiser
6.    Conflict
7.    Niger
8.    Phoenix
9.    Perseus
10.    Plumper
11.    Shearwater
12.    Royalist
13.    Sharpshooter
14.    Nimrod
15.    Roebuck
16.    Assurance
17.    Mohawk
18.    Osprey
19.    Sparrowhawk
20.    Adventure
21.    Fox
22.    Argus
23.    Basilisk
24.    Buzzard
25.    Cyclops
26.    Dragon
27.    Fury
28.    Gladiator
29.    Hecla
30.    Hermes
31.    Inflexible
32.    Rosamond
33.    Sampson
34.    Styx
35.    Vixen
36.    Vulture
37.    Cardoc
38.    Recruit
39.    Triton
40.    Spitfire
41.    Locust

and 70 gunboats

Sunday, 29 November 2015

Seniority in the Maryland Campaign

It seems over at no-one knows the seniority of commanders in the Maryland Campaign. Here are the generals commanding the Army of the Potomac, Departments, Corps and Divisions. Those in [] are not present with the Army, but this is their indicated seniority.

Major Generals
1. George B. McClellan (MG of regs, 14th May '61)  - commanding Army of the Potomac
[Nathaniel P. Banks (MG of vols, 16th May '61) - commanding 12th Corps (not present) and Defences of Washington]
[John A. Dix (MG of vols, 16th May '61) - commanding Dept of Virginia at Ft Monroe]
[Ethan A. Hitchcock (MG of vols, 10th Feb '62) - on Lincoln's staff]
2. Ambrose E. Burnside (MG of vols, 18th Mar '62)
[Franz Sigel (MG of vols, 21st Mar '62) - commanding 11th Corps at Washington]
3. Jesse L. Reno (MG of vols, 26th Apr '62) - commanding 9th Corps until KIA 14th Sept '62
[John G. Parke (MG of vols, 26th Apr'62) - commanding Dept of NC]
4. Edwin V. Sumner (MG of vols, 4th July '62 and BG of regs, 16th Mar '61) - commanding 2nd Corps
[Samuel P. Heintzelman (MG of vols, 4th July '62) - commanding 3rd Corps at Washington]
[Erasmus D. Keyes (MG of vols, 4th July '62) - commanding 4th Corps Yorktown]
5. Fitz John Porter (MG of vols, 4th July '62) - commanding 5th Corps
6. William B. Franklin (MG of vols, 4th July '62) - commanding 6th Corps
7. Joseph Hooker (MG of vols, 4th July '62) - commanding 1st Corps until WIA 17th Sept '62
8. Darius M. Couch (MG of vols, 4th July '62) - commanding division of 6th Corps
9. Israel B. Richardson (MG of vols, 4th July '62) - commanding division of 2nd Corps until MWIA 17th Sept '62
10. Henry W. Slocum (MG of vols, 4th July '62) - commanding division of 6th Corps
[John J. Peck (MG of vols, 4th July '62) - commanding division of 4th Corps at Yorktown]
11. George W. Morell (MG of vols, 4th July '62) - commanding division of 5th Corps
12. William  F. Smith (MG of vols, 4th July '62) - commanding division of 6th Corps
13. John Sedgwick (MG of vols, 4th July '62) - commanding division of 2nd Corps

Top eight Brigadier Generals
14. Joseph K.F. Mansfield (BG of regs by bvt, 4th May '61, with pay as BG starting 14th May '61) - took acting command of 12th Corps until MWIA 17th Sept '62
15. Jacob D. Cox (BG of vols, 17th May '61) - commanding Kanawha division and acting commander 9th Corps after Reno KIA
16. George Meade (BG of vols, 31st Aug '61) - commanding division in 1st Corps and acting commander of 1st Corps after Hooker WIA
17. Oliver O. Howard (BG of vols, 3rd Sept '61) - commanding brigade and commanding division in 2nd Corps after Sedgwick WIA
18. John Newton (BG of vols, 23rd Sept '61) - commanding brigade only
19. Winfield S. Hancock (BG of vols, 23rd Sept '61) - commanding brigade until assigned to command division vice Richardson MWIA
20. George Sykes (BG of vols, 28th Sept '61) - commanding Regular Brigade
21. William H. French (BG of vols, 28th Sept '61) - commanding division in 2nd Corps

Wednesday, 2 September 2015

British Measures of Strength - Winter 1854-5

In a recent discussion on the ever opinionated TFSmith121 had an unfortunate bout of research failure. He threw in some extremely inaccurate numbers for the British Army of the East. Maybe it's worth examining how British returns are filled in.

Here is a link to the actual morning states of the British on various days in the winter of 1854-5 (ref), and here is the data in a nice modern table (click to embiggen), with the same numbers processed into the Federal US equivalents:

(NB: Numbers exclude a brigade of Royal Marines who were at Balaklava with 2nd (Highland) Brigade, 1st Division)

Note the British have more strength categories, and if we'd had some different tables they'd also separate out those Present Under Arms (PUA) into those detailed to the trenches and those available for other duties. We might also separate those "on command" (roughly "extra duty" in US terms), Bâtmen and other employees (roughly "special duty" and "daily duty").

The strength PUA is those available to fight, rather than on other duties. When we read strengths in many British Napoleonic and other histories (like Oman's magnum opus) the "effective strength present" is the total of officers and men PUA, excluding men "on command", sick etc. - see here for an example of Sir John Moore's Army of 1808 (ref). This leads to a distortion, because French strengths are generally reported as roughly "present for duty" and a distortion in favour of British fighting prowess occurs similar to US-CS distortions.

So what of Smith's claim that the British only had 8,000 troops actually present, ignoring the fact that he called the 1st (Guards) Brigade, 1st Division the "Guards Division" and completely missed out the largest division in the army (Light Division)? Simple - he's ignorant of the fact that these numbers, which are perfectly valid and derive ultimately from Kingslake Vol. 6 (ref) are only that portion of the army actually engaged (excluding the Light Division).

Anyone who'd read any regimental histories (or Kingslake) would know that Inkerman was fought by those portions of regiments not assigned to the trenches. Take the 88th Foot (The Connaught Rangers) for example marched to Inkerman with the Grenadier, No. 5, No. 7 and Light companies, and were joined late in the battle by No. 2 company after they'd been relieved by No. 3 company (ref); No.s 1, 4, 6 and 8 companies were on trench duty. The four companies that set out mustered only about 290 men, but this implies 725 effectives in the battalion as a whole. There were 31.2 battalions of infantry (exc/ Marines) with the army and the 3rd November morning state shows 18,498 infantry officers and men PUA for an average of 593 - the 88th may have been a strong battalion, although it is not clear the returns above include the detachment at Balaklava.

As one should note, the total present increased even though the combat strength stayed static due to illness. There was a steady flow of replacements to the old regiments, but 12 full battalions (the equivalent of a full infantry Corps de Armee in 1862 terms) reinforced during the period*.

Here we can see the British had far more sensible measures of strength, but stulti caveant for those who don't understand it!

* Arrival dates:
12th Nov - 62nd from Malta
20th Nov - 97th from Greece
27th Nov - 9th from Malta
4th Dec - 90th from England
9th Dec - 34th from Corfu
15th Dec - 17th from Gibraltar
19th Dec - 89th from Gibraltar
22nd Dec - det. 71st from Corfu
26th Dec - 18th from England
1st Jan - 39th from Corfu
19th Jan - 14th from Malta
3rd Feb - remainder 71st

Tuesday, 25 August 2015

Map: McClellan's Last Campaign

Since I've just got a scan from the Library of Congress of the position of McClellan's army when he was relieved I'll put it up here for future reference. Click to expand.

Sunday, 2 August 2015

Effective vs Present for Duty (1): Army of Northern Virginia

1. Introduction

With an army it is obvious not every single soldier is available as a shooter; that's Heinlein's novel Starship Troopers, not history. The question thus is how many of those present and fit were actually shooters.

Non-shooters (excepting the sick and under arrest) typically fall into three categories, which are only really distinguished by time and whether extra pay is due:

Daily Duty: every morning the company would detail a few men in turn to take their turn assisting the cooks, tending fires, guarding the packs of the rest of the company if going into battle etc.. This was a rolling roster and the privates all took turns.
Special Duty: troops assigned to non-line duties who are not due extra pay (typically for stints of less than 10 days). May be cooks, clerks, teamsters, hospital orderlies etc.
Extra Duty: troops assigned to non-line duties who for greater than 10 days and are due an extra 35 cents p.d.; however they are typically doing the same jobs as S.D. troops, but are generally more or less permanently detached from their regiment.

The troops assigned "daily duty" were always carried in the "present for duty" column in all armies, but the "special" and "extra duty" men sometimes were and sometimes weren't. What's more under pre-war regulations only the men on "extra duty" were broken out into a separate column (a consequence of them drawing extra pay), leaving those on "special duty" mixed in with the PFD.

When looking at numbers with reference to trying to determine the number of shooters it is important to know how to read field or monthly returns.

For the Confederates we have quite a lot of information, and it varies by army.

2a. Army of Northern Virginia

Fortunately Taylor was explicit that at the time of Gettysburg all the "special and extra duty" troops were not counted as "present for duty" (ref). The number of "effectives" is approximately synonymous with the combat strength of the army minus temporary detachments like camp guards and officers. By this period "present for duty" was the number of officers and men for battle, and the "effectives" was the same minus officers.

The question is did Lee ever change definitions, and did Johnston use the same ones?

To answer this all available returns of the Army of Northern Virginia in the OR were entered into a spreadsheet, excluding some divisional returns etc. A large gap should be noted for the spring of 1862.

The 23rd April return for Magruder's corps (ref) was awkward, with McLaws reporting his PFD, aggregate present and present and absent as the same number! For ca. 30th April and 20th May Johnston gives us the approximate "effective strength" of his brigades (noting the second is largely a copy of the first). Thus these returns are excluded, leaving 54 returns from the "Army of Northern Virginia" and 6 returns from 1st Corps, Army of the Potomac. PFD is plotted against aggregate present in figure 1.

Figure 1: Scatter plot of "Present for Duty" vs "Aggregate Present" for the Army of Northern Virginia and 1st Corps, Army of the Potomac. Data from Official Records.

The correlation is remarkable. It is clear that at no point did definitions change, which would be apparent by effectively two different series. The one outlier is the 20th July '62 - the first return after the Seven Days. There is a similar (but not significant) deviation immediately after Gettysburg. The best interpretation seems to be either an underreporting of PFD or a significant breakdown in the army (since wounded were sent to the hospitals around Richmond they were "absent"). This implies that after the losses of the Seven Days and detaching Jackson and Ewell back to the valley Lee still had around 80,000 PFD, ca. 10,000 more than he reported.

However, we can observe differences between Johnston and Lee when they report "effective strength". In both cases it approximates PFD, but Johnston included officers and Lee did not. Whether the transition occurred under Lee or whether Johnston dropped officers from effectives after his 28th February report is unknown. Interpreting Johnston's 30th April and 20th May estimate of effectives is thus problematic. In previous work I have assumed Johnston continued to include officers, but there is little to recommend this other than it being the conservative number.

2b. Army of the Peninsula and Norfolk

Magruder's 31st December '61 return has a PFD:Present ratio of 0.79, lower than the norm in Lee's army. Ergo same deal, non-combatants aren't included in the PFD category. Comparing DR Jones' return of 23rd April (which is AP and P&A only) with his "effectives" one week later gives a ratio of 0.83.

The 31st December '61 and 31st January '62 have both ratios of 0.85, in line with Lee. Interestingly Colston's brigade has PFD 147 offrs and 2,752 other ranks, whereas in Johnston's return of 30th April it has 1,750 effectives; this minor mystery was solved by observing that the earlier returns included 46 artillery pieces emplaced along the James river and their crews.