Wednesday, 2 April 2014

Was Grant outnumbered in May 1863?

Twice in quick secession I have been confronted with the argument that when Grant crossed the Mississippi in May 1863 he was outnumbered. The argument is thus:

Grant should only be accounted for forces that crossed, or even the forces that crossed on the first day, but Pemberton's entire department should be counted.

Lets examine these in order.

1. Grant's strength.

Using the 30th April 1863 returns in the OR (link) we know that Grant's field force consisted of:

73,674 aggregate present and absent, of which:
53,596 are present and
47,152 are "for duty"
with 122 field guns

However, although in the campaign, a division of 13th Corps and some cavalry was left at Helena, Arkansas (4,994 "for duty") and must be deducted. They fought the Battle of Helena in July (link), and thus should be considered if the scope is widened.

Grant crossed the following:

30th April
Army HQ: not reported (and engineers, but not taken into account here)
13th Corps (- District of Arkansas): 17,696 PFD
7th Division, 17th Corps: 6,498 PFD
= 24,194 PFD

1st May
HQ + 3rd Division, 17th Corps: 6,574
= 30,768

It is with these 30,768 men Grant knocks away Bowen's 2 brigade division at Port Gibson.

6-7th May
Main Body, 15th Corps (exc/2nd Division): 11,136
= 41,904 PFD

11th May
2nd Division, 15th Corps: 5,917 (relieved from previous duty by troops from 16th Corps)
= 47,821 PFD

13th May (crosses river)
3rd Division, 17th Corps: 6,498
= 54,319 PFD

In the meantime Grant has already ordered much of 16th Corps up to the Arkansas side of the river, but these don't cross before the field battles so aren't counted. Thus we can estimate Grant had 54,319 Present for Duty east of the Mississippi when advancing on Jackson, although the last division crosses and guards the line of retreat.

However, this excludes cavalry, of which we have no report, and some other small elements.

2. Pemberton's Strength

To construct the argument Grant was outnumbered the practice is to take the strength of Pemberton's whole department, and compare it to only Grant's field force. The direct equivalent return to Grant's is not in the OR, but no new formations join Pemberton so we'll use the 31st March returns (link).

The returns show five military districts, and four divisions. It's worth noting what these districts are:

1st district is at Columbus, Mississippi
2nd district is at Vicksburg
3rd district is as Port Hudson
4th district is at Jackson
5th district is at Grenada

Thus 1st and 5th districts are 130-150 miles north of Jackson, and form a cordon against 16th Corps. 3rd district is 150 miles from Jackson to the SW, and is facing an expedition by Bank's Corps up from New Orleans (numbers from Google Maps routefinder, representing modern road distances).

Thus we can write off 1st, 3rd and 5th districts. They are in no sense in the field around Vicksburg. However, they did send reinforcements to Jackson or Vicksburg, and these do need to be accounted for, as do reinforcements from outside the Department.

2a. Reinforcements to Vicksburg/ Jackson in time for the Battle of Champion Hill

Setting Champion Hill as the vital date, the following reinforcements arrive:

Gregg's Brigade from Port Hudson (2,745 (link)) arrives at Jackson and is the principle formation engaged at Raymond and Jackson. Maxey's brigade (2,747) will follow, but had not sufficient transport to move to Jackson, and did not arrive in theatre until 31st May.

Loring's Division received a reinforcement from Port Hudson - Buford's brigade (2,735), in time for the Champion Hill. It had been detached from Port Hudson and was sent to Bragg's army in April, but recalled.

Waul's Texas Legion was attached to Loring's Division on the 31st March return, but remained at Vicksburg after Loring was cut off (as was the 6th Mississippi Regiment and Loring's sick).

The 5th Mississippi Regiment and 3rd Battalion, State Troops (i.e. militia) were also in Vicksburg, but were small.

One should also note that the two brigades of Maury's division are carried on Smith's and Loring's divisional returns on the 31st March state.

On 2nd May the CSA War Dept ordered Beauregard to send 8-10,000 troops to Mississippi (link). Beauregard sent Walker's and Gist's brigades and protested (link), and will later send another. The heads of these ca. 5,000 men (no report, these are Beauregard's estimates which were higher than Johnston's).

Thus we can estimate that when Grant crossed the Mississippi there were 29,719 PFD in the Vicksburg theatre of operations. Johnston estimates about 6,000 at Jackson with the reinforcements, which seems about right (exclusive of 1,500 under Gist stopped). We can estimate:

Gregg (see above): 2,745
Gist: 2,456 (link)
Walker: 2,379 (under Col. Wilson, prev. link)

= 7,580

2b. Reinforcements reaching Johnston at Jackson after Champion Hill

As noted, Maxey's brigade (2,747) from Port Hudson was en route. Maxey was at Brookhaven on 12th May (60 road miles from Jackson) and aborted, linking up with Johnston later. The brigade was later incorporated into a new division formed under Brig Gen Walker with two other random brigades.

McNair's brigade (1,432 on 20th Feb (link) and Ector's brigade (1,154) were ordered from Bragg's army. They reached Johnston's army about 26th May and were incorporated into the three brigades with him above (2a) to form Walker's division. McNair's brigade was shifted to French's new division 21st June.

Evan's Brigade (1,889 or 2,051 PFD (link, link)) arrived 30th May from South Carolina. It was combined with Maxey's and McNair's brigades to create a new division under Maj Gen French.

Breckinridge's Division arrived 1st June from Bragg's army, well after the field phase. It was 6,143 PFD (link above).

Excluding cavalry (both sides cavalry is excluded as it is difficult to figure) and using the higher figure for Evans we get 13,527 PFD for the forces that reached Johnston after Vicksburg was invested.

2c. Totals

In the Vicksburg defences and covering field forces: 29,719
Johnston's Forces in vicinity of Jackson ca. 14th May: 7,580

Total in the field, including Gist ca. 14-16 May:37,299

Reinforcements received afterwards: 13,527

Grand Total, Confederate Forces in the Campaign: 50,826 (for duty, mixed timings hence estimate)

3. Conclusions

Not only did Grant seriously outnumber the defenders of Jackson and Vicksburg about 5:3 when he launched for Jackson, he in fact moderately outnumbered them and all the reinforcements Johnston received!

In the time Johnston received 13,527, Grant received approximately 28,000 (numbers rough).

The idea that "Grant was outnumbered in theatre" is a nonsense. It is based upon the most extreme apples vs oranges comparisons and worthy of a Jubal Early playing for the Union side.

4. Note on the number of surrendered at Vicksburg

It's not the number of men PFD. It might be the aggregate present and absent or aggregate present because:

a. The Confederate returns say so, 29,376 in the aggregate present and absent for 30th June 1863 (link).

b. Looking at the numbers. The formations that surrendered 4th July 1863 had the following strength at the campaign's start:

PFD: 24,100
Present: 31,133
Present and Absent: 41,227

Now, the campaign has knocked at least 9,000 off each of those numbers, as not many of the casualties were in Loring's division. Now Grant claims 29,491 surrendered (link). Present for duty after casualties would be around 15,000 (indeed early inspection reports show less than 18,000 men defending Vicksburg (link)), although aggregate present could be higher. It deserves further investigation as the number surrendered became a source of contention, and broke down the prisoner exchange cartel.

Saturday, 22 March 2014

Confederate Army in Virginia 1861: McClellan's Odds?

When discussing McClellan eventually the discussion gets round to numbers and estimates. There are usually a number of naive comparisons because commentators fail to check data, and why should they - rigour usually is favourable to McClellan?

McClellan gets a lot of stick from some quarters for his "overestimates" of the enemy strength in Virginia, especially in 1861. His early intelligence of 150,000 or 170,000 available against Washington is derided. However, when doing so no-one bothers to ask the enemies actual strength.

On 30th September 1861 questions raised about where Confederate regiments were led to this table in the OR (link). A few of the unknowns were in Virginia. In total the forces in Virginia were disposed thus:

Manassas: 74 major units (inf regts/ bns or cav regts) and 15 minor units (artillery btys)
Fredericksburg: 9 major units
NW Virginia: 19 major units and 2 minor ones
Kanawha Valley: 11 major units
Winchester: 7 major units (4 of them militia)
Yorktown: 20 major units and 5 minor ones
Norfolk: 18 major units
Richmond: 4 major units
East Shore: 1 major unit

Reducing these to "standard regiments" one gets 165.2 "regiments", a force of ca. 165,000 at full strength. This is rather more than some would suppose.

Of course, we have returns for a bit later, but they're spotty, and there is a lot of confusion about the two corps of the Confederate Army of the Potomac. The actual organisation was:

1st Corps (Beauregard): 1st (van Dorn's) and 3rd (Longstreet's) divisions
2nd Corps (GW Smith): 2nd division (GW Smith's) and the division at Dumfries (Whiting's)
4th Division (EK Smith's) was an independent reserve
Aquia and Valley districts, the new division forming at Leesburg under DH Hill, Jackson's and Loring's divisions in the Valley and the reserve artillery and cavalry and not part of the corps organisation.

Hence we get large variations because different reports don't report all the formations you'd think they did. The most complete state (end of 1861 report, link) shows 76,331 in Johnston's camps (which includes his Valley and Aquia detachments, but not Loring who some months earlier had 11,669). However, in January divisions were still reporting 11-13,000 men present (see the return of 1st Corps, i.e. Van Dorn's and Longstreet's divisions ca. 22nd January 1862, link). Thus a division was equivalent in manpower to a Federal one.

Using Federal equivalents, in September '61 the Confederates had 6 divisions, plus a cavalry brigade and odds and sods around Manassas, another 3 divisions in the Shenandoah and Kanawha valleys, and about 1.5 divisions each at Norfolk and Yorktown. The equivalent of 12 Federal divisions.

On 15th October, McClellan had, excluding the immobile garrisons, 10 divisions and odds and sods (link), with 143,647 along virtually the entire Virginia front (although the Kanawha valley isn't the Potomac divisions responsibility). By November he'd formed 11* divisions with 123,618 aggregate present (inc. reserve cav and arty). Thus if McClellan moved with his entire usable force (remembering in reality he'd have to leave part in Washington, which wasn't yet fortified, with only 67 guns in the fortifications, plus Ft Washington defending the river) he would not have a massive advantage in numbers, although he theoretically would.

The total Confederate force in Virginia ca. December was about 118,306, and they could probably concentrate 90-100,000 men to give battle to McClellan along the Rappahanock line. There is little chance of simply swallowing such a force. McClellan himself estimated he'd need a force of 150,000 for effective operations, exclusive of garrisons. He doesn't appear to have been far wrong...

* Neither Casey's nor Dix's "divisions" were such. "Casey's division" was the training division for new arrivals, and "Dix's division" was the Baltimore garrison. Another such non-field division would be added over the winter and another field division was formed. Before the Peninsula operation Casey's division was filled and taken to the field.

Thursday, 11 October 2012

In response to another blog.

I've just come across a blogspot post of someone who really objects to something I wrote:

I've seen similar arguments a lot from what one commentator once dubbed "ameriteens". One of the fundamental issues I have is a lack of consistency in the argument. Oh, and referencing Howard Fuller. One should always double check some things. For example, he quotes from Fuller as follows:

From 'An Admiralty Return ordered by Parliament on June 30, 1862, "of all Iron-Cased Ships and Floating Batteries Building or Afloat," noted [as of July 17, 1862]:  Iron-hulled [built]                                 Warrior, Black Prince, Defence, Resistance  Iron-hulled Floating Batteries: Trusty, Thunder, Glatton, Aetna, Hector, Valiant, Achilles.  Iron-hulled [building]: Minotaur, Agincourt, Northumberland, Prince Albert  [Data from Clad In Iron, By Howard J. Fuller]

Anyone familiar with the subject in hand would immediately ask where is Royal Oak, Ocean, Prince Consort etc.

One should check the Navy List, the 20th December 1862 edition is handily online: which lists:

  1. Achilles (building at Chatham, reserve from December 1863, Commissioned into the Channel Fleet September 1864)
  2. Agincourt (building at Birkenhead, reserve on 26th May 1864, Commissioned 1867)
  3. Black Prince (in Commission with the Channel Fleet)
  4. Caledonia (in reserve at Woolwich from 2nd February 1863, Commissioned into the Med. Fleet July 1865)
  5. Defence (in Commission with the Channel Fleet)
  6. Enterprise (building at Deptford, Commissioned into the Med. Fleet May 1864)
  7. Erebus (in deep reserve a Portsmouth)
  8. Favourite (building at Deptford, Commissioned into the American Fleet 1866)
  9. Glatton (in deep reserve at Portsmouth)
  10. Hector (building at Glasgow, reserve from October 1862, Commissioned into Channel Fleet 1864)
  11. Minotaur (building at Blackwell, reserve on 15 December 1863, Commissioned into the Channel Fleet 1867)
  12. Northumberland (building at Millwall, reserve in 1866, Commissioned into the Channel Fleet 1868)
  13. Ocean (building at Devonport, in reserve from 23rd March 1863, Commissioned into the Channel Fleet 1866)
  14. Prince Albert (building at Millwall, reserve from 20th May 1864, Commissioned almost immediately into the Channel Fleet to test the turrets)
  15. Prince Consort (building at Pembroke, reserve from 14th January 1863, Commissioned into the Channel Fleet 1864)
  16. Research (building at Pembroke, reserve from March 1864, and Commissioned next month into the Channel Fleet)
  17. Resistance (in Commission with the Channel Fleet)
  18. Royal Alfred (building at Portsmouth, in reserve October 1864, Commissioned into the American Fleet Jan. 1867)
  19. Royal Oak (building at Chatham, in reserve 13th September 1862, Commissioned into 27th April 1863 into the Channel Fleet)
  20. Royal Sovereign (building at Portsmouth, complete as turret ship 20th August 1864 and placed on Harbour Commission in October, but was never fully Commissioned)
  21. Terror (in Commission on the Bermuda station)
  22. Thunder (in deep reserve at Sheerness)
  23. Thunderbolt (in deep reserve on the River Thames)
  24. Trusty (in deep reserve at Woolwich)
  25. Valiant (building at Millwall, reserve from October 1863, Commissioned 1868)
  26. Warrior (in Commission with the Channel Fleet)
  27. Zealous (building at Glasgow, reserve from December 1864, Commissioned into the Pacific Fleet 1866)
Or 12 ships more than the above. Anyway, his direct challenges to me:

  I don't know where 67th Tigers gets his figures, but there were less than ten (10) seagoing ironclads available to the Royal Navy in the years 1861-1865. Perhaps he confused them with less seaworthy craft such as the Aetna-Class monitors, of which four remained afloat after the Crimean War. I do agree with him that the riverine casement ironclads and smaller monitors were incapable of harassing the Royal Navy in a blue water deployment, although some of the bigger monitors would have been capable of ocean travel, but their speed would have been ridiculously slow and coal consumption prohibitive, requiring an escorting collier. The health of the crew under such extended-range circumstances would likewise have been self-defeating. However, as brown water vessels and supported by a comprehensive system of fortifications ashore, the Union Navy would have effectively repelled any Royal Navy effort to occupy a major seaport along the Atlantic coast.
Assuming we ignore the Aetna and Erebus classes still around from 1856 (although it should be noted HMS Terror was active along the US/CS coast) when are 10 seagoing ironclads available even without going to double shifts?

Already in Commission (4): Warrior, Black Prince, Defence, Resistance
Entry to reserve (requires storing, arming and crewing, ca. 1 months work is necessary): Royal Oak (9/62), Hector (10/62), Prince Consort (1/63), Caledonia (2/63), Ocean (3/63), Valiant (10/63)

Thus the "less than ten armoured frigates" becomes invalid in October 1863 when Valiant becomes available. In the event of a war over, say the Trent affair, all these would be in service in 1862. When Valiant becomes available in October 1863 the USN has 11 ironclads (New Ironsides, Roanoke and 9 Passaics), and so by this measure has more. However if the 1856 ironclads are added back in the USN has less, and by losses a few months later the USN will have less anyway.

   As for 67th Tigers' contempt for the casement ironclads, he forgets that the USS Cumberland, while being destroyed by the casement ironclad CSS Virginia, had unloaded salvos of eleven (11) 9" Dahlgren guns from her starboard battery--at point blank range--to no avail. Presumably, the Union casement ironclads would enjoy similar invulnerability, at least during the couple of years a Royal Navy counter-blockade was likely to be brought to bear--and at a pace only as fast as the Palmerston government capacity for making the call on whether or not to declare war on the United States and support the Confederacy in defense of slavery!
USS Cumberland of course did quite a bit of damage to CSS Virginia, with her 12 engaged guns knocking out 2 of Virginias. Cumberlands problem was that she was a sail ship at anchor in the age of steam.

 Regarding 'although about this time the RN starts putting ordnance into service capable of piercing Monitor turrets at extreme ranges'. Range was worthless without a seaworthy delivery system. Range was worthless without a competitive arc of fire, given the ability of the Monitor type to turn perpendicular to a serious threat and, thus, minimize the enemy's target zone, blunting the benefit of traditional broadsides volley fire. I doubt improved shot fired from exiting Royal Navy guns could have sufficiently zeroed in on the next-to-invisible silhouette of Monitors at 'extreme range'. I mean no disrespect to 67th Tigers, but I wonder whether he has taken these factors into account. At, under Turrets, it states: "The Royal Navy began using turreted guns in 1864. Seagoing Royal Navy turret ships of the 1860s had masts and rigging, a protective forecastle, and a poop on the stern, which together limited the arc of fire to 120 to 132 degrees. The weight of the hull armor, turrets, and masts made these ships dangerously top heavy [Naval Guns And Gunnery, "Turrets"].
The period was one of flux. In 1862 RN ships mounted a mix of guns. The armoured ships mainly mounted the 68 pdr 95 cwt which was the best armour piercing piece available, and has a significantly better performance against armour than any gun afloat except perhaps the 15" Dahlgren after heavy charges are authorised. Unlike the USN the RN issued steel shot for their guns to improve performance against armour. A 72 lb steel shot propelled by 16 lbs of naval powder from the 68 is a significantly better proposition against armour than the 11" Dahlgren.

The Armstrong 7" RBL was primarily for use against wooden ships. Their accurate shellfire (as opposed to the high velocity punching power of the 68 pdr) was more dangerous to wooden ships. Solid shot from this piece would still damage a monitor though.

However, things moved on. By the time some of the latter heavy monitors are complete the RN have adopted a high velocity muzzle loading rifle and palliser shot and shell. The 7" RML of 1864 could penetrate 7.7 inches of wrought iron plating, and hence would break straight through a typical monitors turret (remembering lamination effects).

  The last traditional ships of the line built for the Royal Navy was the Duncan-Class and the Bulwark-Class, which actually was a third sistership. They sported 101-guns each, mounting a heavy battery of 36 8" naval rifles on the gun deck--8" being too inferior a caliber to penetrate the Monitor's turret. One of the Duncan's completed as the single-ship Bulwark-Class, which offered no improvement in naval gunnery capable of penetrating those troublesome little Monitors.
No Bulwark was completed as a wooden "line of battle ship". All were completed as ironclads.

  The four Prince Consort-Class iron clads that would also have been available during the Civil War for Royal Navy use would have been splendid opponents for the wooden Union Navy, but, again, with only 24 7" guns, would have managed only to rattle the crew inside the Monitors, although such a psychological turmoil, coupled with the poor ventilation when the turret was closed up for battle stations, might well have limited the time a Monitor could spend in any single engagement. Consider, too, that the traditional broadside arrangement of the Warrior and the Prince Consort's meant only half of their weaponry could be brought to bear on a target at any given time. Twelve (12) 7" guns being fired at the shadowy Monitor silhouette, at far range, were not likely to score a salvo hit, but perhaps one or two might have struck, bouncing off.
In 1863 they would have had 68 pdrs as their main armament.

  Regarding the Resistance and the Defence, we have again an inadequate caliber of naval gunnery. After 1867, the two ships mounted just two (2) 8" and fourteen (14) 7" breech-loading naval guns. 
Muzzle loaders on the Woolwich principle. The 7" and 8" will be putting shell through most monitor turrets at ~ 800 yars.

  The Armstrong guns.
  As far as I can tell, the older 68-pounder smooth bore was still the Royal Navy's best weapon for use against a Civil War monitor. Armstrong guns featured a breech loading mechanism that could not endure the explosion of heavy cartridges, so lighter cartridge charges were used, resulting in a lower muzzle velocity--and a diminishment in armor-penetrating impact. No Armstrong gun available to the Royal Navy in time for a notional deployment against the Union Navy could penetrate the armor thickness of HMS Warrior. The 68-pounder, however, was able to penetrate such armor.

When? In 1862 the 68 pdr with steel shot was used against armour. By 1864-5 muzzleloading rifles with chilled iron shells (i.e. armour piercing and with a bursting charge) were in use.

  'However, the lack of such ordnance meant only one per ship could be supplied....Against armored vessels, only the 15" matters....' 67th Tiger's takes a little sidestep away from the obvious fact that an insufficient number of British ironclads capable of surviving a transatlantic voyage existed to overcome the numbers of Union ironclads capable of resisting Royal Navy woodies. HMS Warrior had only just commissioned in August, 1861, and fired a conventional broadsides of 68-pounders of the sort deployed in the Crimean War and used aboard the Aetna-Class monitors as well. Bear in mind, that the CSS Virginia had been equipped with six 9" Dahlgren smoothbore naval guns and none of these shells had been successful in the fight with the Monitor--at close range, no less. So the existing Royal Navy broadsides would have been feeble against even the first generation of Monitors and, considering the low profile of the "cheesebox on a raft", most of the shells fired in these broadsides would have splashed harmlessly around their ironclad targets
Here you state a correct fact but don't get the import. Virginia had shells, not shot (and certainly not armour-piercing shot). If she'd had a supply of shot Monitor would probably have gone to the bottom. Look at the effect of the Charleston forts guns on the Monitors and quite long ranges. The Confederate gunners hit the Monitors about 20% of the time at ~ 1,000 yards.

If "surviving a trans-atlantic voyage" is the parameter then it's all of the ships mentioned above. HMS Terror had a successful career in American waters.

However, even wooden ships can fight ironclads and win (see SMS Kaiser at Lissa). If a Monitor was exposed to the concentrated firepower of a 2-decker it would not last long. In the 15 minutes it would take a Monitor crew to get a shot off with both guns about 700 shots would come back. The sheer weight of fire, with hundreds of hits received per one returned, would break plates, jam the turret, smash the gun shutters and muzzles etc., and the wooden ship has one or two holes (and if unlucky shells exploding). The Monitor would be overwhelmed and sunk eventually.

  The fact that the Monitor type was much more maneuverable was made manifest in the debut of the iron clad at Hampton Roads, Virginia. In Scrutinizing Naval Warfare During the Civil War: The Ironclad, the unposted author writes, "The Virginia's captain ignored what at first appeared to be a floating piece of iron junk. But as the Monitor slipped close by, rotated its turret and fired a near point-blank shot, it became obvious that the Virginia had a fight on its hands. The Monitor was far more maneuverable than the Virginia, and the two circled each other, blasting away, neither causing any noticeable damage to the otherBecause the guns didn't seem to work, each attempted unsuccessfully to ram and disable the other ship. As the battle went on, both ships experienced a host of mechanical problems, but the Virginia had the most trouble, leaking badly and displaying a noticeable crack in her armor. The Virginia did get a shot off at the pilothouse, severely wounding the pilot and causing the Monitor to drift away uncontrolled. Thinking she had won a decisive victory, the Virginia made her way slowly toward home. By the time the Monitor's officers regained control, the Virginia had steamed away. After more than four hours of fighting, both ships were glad to be done for the day, each believing it had won" [].
Oddly the unwieldy Warrior turns quicker. Her turning circle is about twice as wide as Monitors, but she moves around it at nearly 4 times the speed.

  Regarding 'the Passaic is not a competitor for most European ironclads', we are once again faced with 67th Tigers' refusal to acknowledge the paucity of European ironclads in the first place. Anyway, where the Gloire could steam at 11 knots and theWarrior at 14.5 knots, in a close-in engagement, featuring ranges that would enable the 68-pounders to be brought into action, the smaller Monitors could outmaneuver the big ships and thwart the latter's ability to bring their full broadsides to bear. A Union skipper was not likely to simply wait for the line of battle of form-up! The limited number of Royal Navy ironclads available for use in 1863 would have been able to operate together and box-in their targets, but their targets would have been proportionately reduced, cancelling out any strategic impact on the overall progress of the war. At best, the Royal Navy could command a single harbor, but there were many harbors  The issue of converting steam frigates to iron clads brings the USS Roanoke to mind. Her seakeeping qualities had been dangerously curtailed with the redesign and she had been restricted to harbor defense. Presumably, the Royal Navy, too, would have to overcome similar problems in the conversion of their wooden-hulled steam frigates, as well.

How does a 4 knotter choose the range when fighting a 14 knotter?

Just because the USN screwed up Roanoke doesn't mean the RN screwed up Royal Oak etc., nor would they screw up the planned plating of large numbers of ships if war came. They could plate a frigate in a month if necessary, but never needed too.

Worryingly this looks like thesis lead research.

Wednesday, 12 September 2012

Some more on the signals at Glendale

During the 1864 election campaign there was a dispute about whether McClellan was on the armoured corvette USS Galena during the last three days of the Seven Days Battles (30th June, 1st and 2nd July). Of course neither the extreme Democrats (who tried to argue he wasn't on the Galena) nor extreme Republicans (who tried to argue he spent the whole period on the Galena and contributed nothing to the campaign) were right.

Now of course it's accepted he spent a lot of time onboard USS Galena. However the notion that he did not influence the battle from there is questionable. Stephen Sears denie McClellan had any telegraphic communications with his army and believes McClellan had "lost the courage to command" (Gates of Richmond pgs 280-1). Of course he is utterly wrong in this as Lt Clum's signals detachment accompanied McClellan and kept him in contact with his army. One must speculate how he missed a lot of evidence that completely undermines his argument.

The evidence of the assistant surgeon of USS Galena, himself apparently quite an extreme radical Republican gives some detail about the signals. See

One of the most important pieces in this testimony is what the message that brought McClellan ashore. There were two messages close together reading:

"McCall is breaking"


"Sumner is having a hard time."

These messages (apparently from Porter) caused McClellan it immediately return to shore and make a beeline for Sumner. By the time he arrived Sumner and Heintzelmann had moved forces and had repelled Longstreet.

The timings of McClellan's movements are confused in the literature. Even Ethan Rafuse (considered by some to be a "McClellan supporter", something I don't really agree with) suggests McClellan didn't arrive onshore until 9-10pm, which given Heintzelmann's 8pm meeting with McClellan seems to be much later than he came ashore.

On thing that is notable is that the next time McClellan had to go to a meeting with Rodgers he sent a very specific note to Sumner that he was in command due to seniority. Why? I would hazard he was unhappy that Sumner didn't "step up" as he automatically should (see Sumner's testimony to the JCCW: ) and spelt out that as senior Corps Commander he was also 2i/c, although responsibility seems to have been something Sumner made a virtue out of avoiding (just as Franklin and Burnside made a virtue out of ignoring orders).

More work needs doing on McClellan's movements that day, and what signals he sent and received from the Galena. I suspect a very different picture that Sears et al. will prove to be correct.

Dan Vermilya - Harsh Memorial Scholar

Dan Vermilya, this years Harsh Memorial Scholar recently presented his research on the strength of the Federal Army at Antietam. It's going to be published as a paper and, I'm happy to say, his conclusions are broadly the same as mine. Can't wait to see it.

His blog is excellent. See for his take on the Federal strength.

Tuesday, 18 October 2011

Johnston's Strength around Richmond in 1862

I recently purchased a copy of Steven Newton's Joseph E. Johnstonand the Defence of Richmond (amazon) - his PhD thesis.

Appendix B I found fascinating, which was an attempt to derive Johnston's actual strengths, and Newton calculates a much larger force than Johnston is typically credited with. On the Peninsula his starting point is the almost complete 30th April trimonthly field return (there should be returns for the 10th and 20th April, but they seem not to have survived). Newton notes the figure given is for "effectives" and excludes a few organisations. Using Livermore's conversion (which is simply to add back in officers usually) Newton estimates a strength on 30th April of 65,015 PFD. Given known disease rates Newton calculates Johnston arrived on the Peninsula with roughly 72,739 PFD. This is a significantly higher figure than usually supposed.

More impressive is Johnston's force at Seven Pines, which is calculated at 87,890 PFD on 31st May. With reinforcements that arrived the next day the field force would be 94,813 PFD, and this excludes the heavy artillery in the Richmond fortifications and other posts.

This makes Lee's strength of 113,282 (fn 75 to chapter 10 of Harsh, McClellan's War, referencing LW Tenney's MA thesis) during the Seven Days more credible. It always seemed odd the sudden apparent jump in strength under Lee. It now seems likely the strength increase was far more moderate.

Tuesday, 3 May 2011

Ballistics redux

One thing I've noticed is that the ACW has much less lethal musketry than earlier wars. A typical mid or late war battle produces 6 or 7 wounded for every man killed, whilst early war battles and older Napoleonic battles often produce 2 or 3 wounded per man killed. There is an explanation. The new rifled small arms in use are in fact lower energy than the old smoothbore muskets. To wit I calculate:

Figure 1: kinetic energy of an M1842 smoothbore round (.65 ball, ballistic coefficient 0.095, muzzle velocity 1,500 fps) vs an M1861 rifle-musket round (.5775 Burton ball, ballistic coefficient .160, muzzle energy 963 fps).

My *very* rough calculations suggest that roughly 90% of those shot with a Springfield or Enfield type weapon survived the wound, which seems to be an improvement over the maybe 75% who survived smoothbore balls. The problem is that our statistics are skewed, as those killed outright never got to the hospitals to make it into the statistics (the same happened to the killed and wounded in the Crimea, with a doctor reporting only 6% of those he saw were shot in the chest (ref)).