Analysis of One Thousand Union Infantry Casualties at Gettysburg

Tom Elmore

Captain
Member of the Year
Joined
Jan 16, 2015
Having recently identified (by name and unit) 1,000 Union infantrymen whose type of wound (or other injury) at Gettysburg was recorded, a straightforward percentage breakdown is presented below:

68.7 % - struck by gunshot alone, 687 cases (see footnote #1)

25.2 % - struck by artillery (direct) alone, 252 (footnote #2)
1.3% - struck by another object impelled by artillery (indirect), 13 (footnote #3).

2.4 % - struck by artillery and gunshot, 24 (footnote #4)

0.2 % - struck by gunshot and a clubbed or butt end of a musket, 2
0.2 % - struck by gunshot and a bayonet, 2
0.1 % - struck by gunshot and a saber, 1
0.1 % - struck by clubbed musket alone, 1
0.1 % - struck by bayonet alone, 1
0.1 % - struck by saber alone, 1

1.6 % - injuries from all other causes, 16 (footnote #5)

Footnotes:
#1 – at least 35 (five percent) of the 687 cases involved multiple (two or more) gunshots. Also, in three separate cases two men were wounded by the same bullet.
#2 – includes five likely instances of “friendly” artillery fire.
#3 – sheer momentum sent a lot of intermediate objects flying, like fence rails, or fragments from stone walls and outcroppings. Additional casualties were inflicted by detonated caissons, limbers and certain cartridge boxes.
#4 – some of these 24 cases involved contradictory information, or else the lead cause for a disabling wound could not be determined. An even split between gunshot and artillery as the primary cause seems reasonable.
#5 – among the 16 examples, six involved a horse, whether it was falling, bucking or running loose. Four other cases were likely linked to artillery fire – a broken tree limb, bricks knocked loose from a building and stones thrown from an unidentified source. Two involved a rout – once when an officer used his sword on a fleeing man, and another time when a fleeing soldier used his gun butt on an officer. Other cases run the gamut, including a man falling into the railroad cut, a ramrod injury and even a soldier struck by a comrade who was acting out a nightmare.

Other comments:
-The data shows gunshot as the cause for seven in ten of the Union infantry casualties, while artillery felled at least one out of every four infantrymen. By comparison, percentages for the Confederate infantry (1,857 individuals) casualties are: gunshot – 78.8 %; artillery (direct and indirect) – 18.7 %; gunshot and artillery – 1.9 %; all other causes – 0.6 %.
-I have grouped the apparent examples of hand-to-hand combat. The numbers reveal that the attention we accord to close quarters combat far exceeds its actual impact. That includes the vaunted bayonet, which probably saw far more use as a candle holder, or to fix a gun stock in the ground to hold up a shelter tent, than it did as an effective battlefield weapon.
 
Colonel Trevor Dupuy's overall assessment from his HANDBOOK ON GROUND. FORCES ATTRITION IN MODERN WARFARE, SEPTEMBER 1986 lists the following casualty statistics for 19th Century warfare. The majority of the post-1860 stats were of course compiled from the American Civil War. So you can see where you would expect to see casualties from hand-to-hand fighting (saber and bayonet), only 4-6% of the overall casualties were a result. Infantry small arms at 85-90%, and artillery at 8-10%.

Figure 18, page 55
Causes of Battlefield Casualties in the 19th Century
Artillery
Before 1850: 40-50% After 1860: 8-10%

Infantry Small Arms
Before 1850: 30-40% After 1860: 85-90%

Saber and Bayonet
Before 1850: 15-20% After 1860: 4-6%

Significantly higher than average is the Gettysburg artillery percentage at over 25%, is much closer to the casualty percentage expected pre-1850 from artillery. Correspondingly, small arms casualties are significantly lower than the average at under 70%. Also lower are the saber and bayonet casualties. I've always liked statistical analysis, even when grossly aggregated, because they give you a baseline from which to compare individual battles for specific deviations.

@Tom Elmore your Gettysburg analysis is very interesting in that respect. For instance, can we derive any specific rationale for why this particular group of Union infantry suffered significantly higher artillery casualties and fewer small arms casualties than the "norm?" Is this "pool" of infantry truly a "random sample" so that we could apply the statistical results across the entire Army of the Potomac infantry at Gettysburg? Did the casualty percentages vary significantly from day to day and if so, why? Those would be just a few of the questions I would have. I'm sure we could think of many more. Thanks for posting!
 
Very interesting. How did you pick the 1000?
I've just been accumulating the data over time and finally reached 1,000. So all seven corps are well represented, which makes a difference. For instance, the Twelfth Corps fought largely in the woods on Culp's Hill, and as a result had significantly less artillery casualties (roughly 10 percent).
 
Colonel Trevor Dupuy's overall assessment from his HANDBOOK ON GROUND. FORCES ATTRITION IN MODERN WARFARE, SEPTEMBER 1986 lists the following casualty statistics for 19th Century warfare. The majority of the post-1860 stats were of course compiled from the American Civil War. So you can see where you would expect to see casualties from hand-to-hand fighting (saber and bayonet), only 4-6% of the overall casualties were a result. Infantry small arms at 85-90%, and artillery at 8-10%.

Figure 18, page 55
Causes of Battlefield Casualties in the 19th Century
Artillery
Before 1850: 40-50% After 1860: 8-10%

Infantry Small Arms
Before 1850: 30-40% After 1860: 85-90%

Saber and Bayonet
Before 1850: 15-20% After 1860: 4-6%

Significantly higher than average is the Gettysburg artillery percentage at over 25%, is much closer to the casualty percentage expected pre-1850 from artillery. Correspondingly, small arms casualties are significantly lower than the average at under 70%. Also lower are the saber and bayonet casualties. I've always liked statistical analysis, even when grossly aggregated, because they give you a baseline from which to compare individual battles for specific deviations.

@Tom Elmore your Gettysburg analysis is very interesting in that respect. For instance, can we derive any specific rationale for why this particular group of Union infantry suffered significantly higher artillery casualties and fewer small arms casualties than the "norm?" Is this "pool" of infantry truly a "random sample" so that we could apply the statistical results across the entire Army of the Potomac infantry at Gettysburg? Did the casualty percentages vary significantly from day to day and if so, why? Those would be just a few of the questions I would have. I'm sure we could think of many more. Thanks for posting!
My guess is that Gettysburg stands out from other battles fought in Virginia because the ground is much more open and thus more advantageous to the artillery arm, both from the standpoint of visibility and the fact that artillery rounds do not penetrate far into a wooded area.
 
First of all, well done! What a painstaking project!

Do the sources you are using report only wounds to Union soldiers, or are wounded Confederates treated by Union doctors also included?

I assume that the 1000 are all soldiers who were brought into an aid station or field hospital. It would seem unlikely that there would be any way to track what type of weapon inflicted fatal wounds for those who died on the field. At least such would not be in sufficient numbers to make an analysis worthwhile. One would suspect that most KIA wounds that are identified as to weapon would be officers, which would lead to a skewed set of data.

Back when my sons were Boy Scouts, the guide book they used for Gettysburg stated that there were no bayonet wounds, which of course is incorrect. Some military writers of the time period suggested that the value of the bayonet was its psychological effect. If true, then rather than present as a wound, the effect would b for the enemy to flee or surrender. Accordingly, it may be that at least some of the bayonet casualties were POW's.

Occasionally, someone will argue that the lower numbers of artillery wounds reported demonstrated that the artillery of the day was largely ineffective. The typical counter-argument is that many of the artillery inflicted casualties were killed rather than wounded. There is also some argument that artillery is important for its psychological effect. However, one would be hard pressed to find an example at Gettysburg of an assault that was deterred or a defensive position abandoned because of artillery.
 
First of all, well done! What a painstaking project!

Do the sources you are using report only wounds to Union soldiers, or are wounded Confederates treated by Union doctors also included?

I assume that the 1000 are all soldiers who were brought into an aid station or field hospital. It would seem unlikely that there would be any way to track what type of weapon inflicted fatal wounds for those who died on the field. At least such would not be in sufficient numbers to make an analysis worthwhile. One would suspect that most KIA wounds that are identified as to weapon would be officers, which would lead to a skewed set of data.

Back when my sons were Boy Scouts, the guide book they used for Gettysburg stated that there were no bayonet wounds, which of course is incorrect. Some military writers of the time period suggested that the value of the bayonet was its psychological effect. If true, then rather than present as a wound, the effect would b for the enemy to flee or surrender. Accordingly, it may be that at least some of the bayonet casualties were POW's.

Occasionally, someone will argue that the lower numbers of artillery wounds reported demonstrated that the artillery of the day was largely ineffective. The typical counter-argument is that many of the artillery inflicted casualties were killed rather than wounded. There is also some argument that artillery is important for its psychological effect. However, one would be hard pressed to find an example at Gettysburg of an assault that was deterred or a defensive position abandoned because of artillery.
The descriptions come from many first-hand accounts, while a sizeable number were cases from The Medical and Surgical History of the War of the Rebellion. I have wondered about how the data might be inadvertently skewed. For instance, artillery solid shot was usually fatal, but it was usually intended for counter-battery fire at long distances. Adjacent infantry who absorbed such strikes due their close proximity to friendly artillery were either waiting to charge or were passive defenders, and thus had time to record what they saw, so perhaps some over-reporting arises. Once engaged in a pitched battle they were fully occupied with the task at hand. Now we read of devastating canister blasts at close range, the effects of which were likely underreported. We also have many next-of-kin to reliably report what happened to their relations. I think it comes mainly down to sample size, which determines the margin of error, and with more data comes more confidence in the results. To this point, when I had about 1,000 Confederate cases, I saw about one in four were caused by artillery, but now that I have 1,800+ examples, it has dwindled to roughly one in five. What really surprises me is that it is a lower percentage than the Union sample, because we often hear about the superiority of the Union artillery as compared to their Confederate counterparts. Considering that artillery effectiveness drops off as the range increases, intuitively the attacker (normally the Confederates at Gettysburg) should experience greater losses from enemy artillery given the amount of open ground to be traversed - like the major battleground between Seminary and Cemetery ridges.
 
Great Info. I'm interested as well as how did you pick the 1000 soldiers?
 
Having recently identified (by name and unit) 1,000 Union infantrymen whose type of wound (or other injury) at Gettysburg was recorded, a straightforward percentage breakdown is presented below:

68.7 % - struck by gunshot alone, 687 cases (see footnote #1)

25.2 % - struck by artillery (direct) alone, 252 (footnote #2)
1.3% - struck by another object impelled by artillery (indirect), 13 (footnote #3).

2.4 % - struck by artillery and gunshot, 24 (footnote #4)

0.2 % - struck by gunshot and a clubbed or butt end of a musket, 2
0.2 % - struck by gunshot and a bayonet, 2
0.1 % - struck by gunshot and a saber, 1
0.1 % - struck by clubbed musket alone, 1
0.1 % - struck by bayonet alone, 1
0.1 % - struck by saber alone, 1

1.6 % - injuries from all other causes, 16 (footnote #5)

Footnotes:
#1 – at least 35 (five percent) of the 687 cases involved multiple (two or more) gunshots. Also, in three separate cases two men were wounded by the same bullet.
#2 – includes five likely instances of “friendly” artillery fire.
#3 – sheer momentum sent a lot of intermediate objects flying, like fence rails, or fragments from stone walls and outcroppings. Additional casualties were inflicted by detonated caissons, limbers and certain cartridge boxes.
#4 – some of these 24 cases involved contradictory information, or else the lead cause for a disabling wound could not be determined. An even split between gunshot and artillery as the primary cause seems reasonable.
#5 – among the 16 examples, six involved a horse, whether it was falling, bucking or running loose. Four other cases were likely linked to artillery fire – a broken tree limb, bricks knocked loose from a building and stones thrown from an unidentified source. Two involved a rout – once when an officer used his sword on a fleeing man, and another time when a fleeing soldier used his gun butt on an officer. Other cases run the gamut, including a man falling into the railroad cut, a ramrod injury and even a soldier struck by a comrade who was acting out a nightmare.

Other comments:
-The data shows gunshot as the cause for seven in ten of the Union infantry casualties, while artillery felled at least one out of every four infantrymen. By comparison, percentages for the Confederate infantry (1,857 individuals) casualties are: gunshot – 78.8 %; artillery (direct and indirect) – 18.7 %; gunshot and artillery – 1.9 %; all other causes – 0.6 %.
-I have grouped the apparent examples of hand-to-hand combat. The numbers reveal that the attention we accord to close quarters combat far exceeds its actual impact. That includes the vaunted bayonet, which probably saw far more use as a candle holder, or to fix a gun stock in the ground to hold up a shelter tent, than it did as an effective battlefield weapon.
Amazing stuff!

To build off of your point about the bayonet. I'm a little more than halfway through the medical records for the regiment I'm working on. In ~5,000 compiled records, I have found one entry of a bayonet wound. It appears that, at least for the 147th New York, picket duty and camp axes were more dangerous than the bayonet.
 
Amazing stuff!

To build off of your point about the bayonet. I'm a little more than halfway through the medical records for the regiment I'm working on. In ~5,000 compiled records, I have found one entry of a bayonet wound. It appears that, at least for the 147th New York, picket duty and camp axes were more dangerous than the bayonet.
By the way, as an example, these are entries for the 147th New York on my list (W for wounded, K for killed):

Miller, Francis C., LCol, 147 NY, 1 July, W bullet grazed his scalp (GS)
Box, John F., 2Lt, A/147 NY, 3 July, W arm broken by a ball, amputated at shoulder (GS)
Hallock, Victor B., Pvt, H/147 NY, 1 July, W ball through muscle and cords of leg (GS)
Sperry, Edwin M., Sgt, C/147 NY, 1 July, W slightly shell fragment on ankle (A)
Taylor, Sylvester J., 2Lt, E/147 NY, 3 July, K by musket ball before he returned with cartridges (GS)

Summary: Artillery-1, Gunshot-4

Edit: One more to add, see: https://civilwartalk.com/threads/ca...is-comrade-at-gettysburg.161217/#post-2110905
 
Last edited:
This afternoon I was reading Volume 27, part 2, page 487, and came across Saml. McD. Tate's letter to Zebulun Vance on July 8, 1863. It interested me for a couple of reasons, one was his account of the action on July 2nd. He was a Major commanding the Sixth North Carolina on the north side of Cemetery Ridge I believe, and relates the total occupation of a gun emplacement but due to no support, it was abandoned. Here is the near end of that report;

"Inside the works the enemy were left lying in great heaps, and most all with bayonet wounds and many with skulls broken with the breeches of our guns. We left not a living man on the hill of our enemy. I write this now for fear I will not live to write at leisure hereafter."

I have heard many comment then during the war of so few bayonet wounds, and wanted to bring particular notice of Tate's report, also wondering if others may wish to comment. Maybe this was an isolated incident?
Lubliner.
 
This afternoon I was reading Volume 27, part 2, page 487, and came across Saml. McD. Tate's letter to Zebulun Vance on July 8, 1863. It interested me for a couple of reasons, one was his account of the action on July 2nd. He was a Major commanding the Sixth North Carolina on the north side of Cemetery Ridge I believe, and relates the total occupation of a gun emplacement but due to no support, it was abandoned. Here is the near end of that report;

"Inside the works the enemy were left lying in great heaps, and most all with bayonet wounds and many with skulls broken with the breeches of our guns. We left not a living man on the hill of our enemy. I write this now for fear I will not live to write at leisure hereafter."

I have heard many comment then during the war of so few bayonet wounds, and wanted to bring particular notice of Tate's report, also wondering if others may wish to comment. Maybe this was an isolated incident?
Lubliner.

Bayonet wounds were likely fairly rare but the reason that they just aren't documented is that those who suffered them usually died fairly quickly (if not immediately) and thus would not be known in the medical records. The only way that we know about most of them come from participants' writings rather than official documents. After all, nobody was examining the bodies of the thousands of dead on the battlefield to determine how they were killed.

Ryan
 
This afternoon I was reading Volume 27, part 2, page 487, and came across Saml. McD. Tate's letter to Zebulun Vance on July 8, 1863. It interested me for a couple of reasons, one was his account of the action on July 2nd. He was a Major commanding the Sixth North Carolina on the north side of Cemetery Ridge I believe, and relates the total occupation of a gun emplacement but due to no support, it was abandoned. Here is the near end of that report;

"Inside the works the enemy were left lying in great heaps, and most all with bayonet wounds and many with skulls broken with the breeches of our guns. We left not a living man on the hill of our enemy. I write this now for fear I will not live to write at leisure hereafter."

I have heard many comment then during the war of so few bayonet wounds, and wanted to bring particular notice of Tate's report, also wondering if others may wish to comment. Maybe this was an isolated incident?
Lubliner.
The fight on Cemetery Hill the night of July 2 was particularly ferocious - it was dark, the ground was rough, it was a critical position that Union men knew had to be held, it was held by Union troops of uneven quality, and fighting involved artillery crews, who if armed, did not have long-range personal weapons.

Many of these factors were uncommon - the night fighting, the artillery standing their ground up close instead of trying to haul off, the fighting with a physical objective in mind.

I pulled Pfanz's Culp's Hill and Cemtery Hill off of the shelf and there are numerous accounts of hand-to-hand combat in this particular fight. Not sure if the quote you mention was used, but Pfanz does cite Tate.

While the argument of this thread is that melee wounds were more rare than one would think, they did occur - especially in unique circumstances.
 
Last edited:
I think one conclusion from this study has to be that the source of information does not include most of those who died quickly, since they were not carried to the medical tents to be tended and recorded. There would be plenty of dead from both artillery and rifle fire in this uncounted group, with canister and solid shot deaths being largely uncounted.
 
I must admit I was misled by the idea that soldiers rarely used their bayonets, which didn't make sense to me when every assault would begin with the weapon in place. I also noticed the use of the rifle breech to club the enemy with more than the stock or butt-plate. Likewise I read of a pistol-whip that struck the enemy with the barrel and not the handle, leaving a scar and knocking unconscious. Unless the enemy was already flat on the ground where the bayonet could be driven through him, I can see how one sitting could be knocked senseless by the breech when quickness is necessary. That did clear my mind concerning the number of wounds made by the bayonet, but the original statement of uncommon usage might be exaggerated. Thanks to all that replied.
Lubliner.
 
Something I just thought of, and I wanted to add it because I am enjoying this conversation so much.

The ratio of bayonet / clubbing wounds to bullet / artillery wounds can also be explained by considering volume. An infantry regiment could take any number of positions (up right in the open, prone, on a knee, behind trees, behind breastworks) and fire for as long as they had ammunition. While estimates suggests only 1 in 100 balls fired hit home during the war, a regiment of 400 men firing 40 rounds a piece could likely cause 100 casualties over the course of an hour. Units standing 100 or 150 yards apart and blasting away at each other is where many of the shocking casualty counts of the war come from.

Melee is short - it's exhausting, it's not cohesive, and it's all about momentum. Sustained hand-to-hand combat wasn't feasible, especially among men who had limited training and experience in such scenarios. This is certain to have impacted the number of melee-type wounds.
 
Back
Top