Lee's Army during the Overland Campaign

A Numerical Study

Alfred C. Young III

• Publisher: LSU Press
• Year: 2013
• Print ISBN:
  9780807151723
• eBook ISBN:
  9780807151747
• Topic: History
• Subtopic: American Civil War History
• Index: History

The initial confrontation between Union general Ulysses S. Grant and Confederate general Robert E. Lee in Virginia during the Overland Campaign has not until recently received the same degree of scrutiny as other Civil War battles. The first round of combat between the two renowned generals spanned about six weeks in May and early June 1864. The major skirmishes—Wilderness, Spotsylvania, and Cold Harbor—rivaled any other key engagement in the war. While the strength and casualties in Grant's army remain uncontested, historians know much less about Lee's army. Nonetheless, the prevailing narrative depicts Confederates as outstripped nearly two to one, and portrays Grant suffering losses at a rate nearly double that of Lee. As a result, most Civil War scholars contend that the campaign proved a clear numerical victory for Lee but a tactical triumph for Grant.

Questions about the power of Lee's army stem mainly from poor record keeping by the Confederates as well as an inordinate number of missing or lost battle reports. The complexity of the Overland Campaign, which consisted of several smaller engagements in addition to the three main clashes, led to considerable historic uncertainty regarding Lee's army. Significant doubts persist about the army's capability at the commencement of the drive, the amount of reinforcements received, and the total of casualties sustained during the entire campaign and at each of the major battles.

In Lee's Army during the Overland Campaign, Alfred C. Young III addresses this deficiency by providing for the first time accurate information regarding the Confederate side throughout the conflict. The results challenge prevailing assumptions, showing clearly that Lee's army stood far larger in strength and size and suffered considerably higher casualties than previously believed.

PART ONE

Lee’s Army: An Overview

CHAPTER I

The Initial Strength of the Army of Northern Virginia

The typical historical account of the campaigns of 1864 and 1865 in Virginia has the forces of Lee’s Army of Northern Virginia getting the most out of limited, if not meager, manpower, armaments, commissary supplies, and other resources before finally being overcome by Grant’s numerically superior and significantly better equipped and supplied Army of the Potomac. Regarding these meager manpower resources, historians have generally concluded that Lee’s army entered the Wilderness on May 5, 1864, with about one-half of the numerical strength of Grant’s army.

Estimates for the Confederate army at the commencement of the campaign conventionally have ranged from about 60,000 to 62,000 men. The primary source for this information is the “Abstract of Field Returns for the Army of Northern Virginia,” dated April 20, 1864, found on pages 1097–1098 of volume 33, series 1 of the Official Records. This is the last available return or muster roll for the Confederate forces prior to May 5. The units listed as present for duty in this return are shown in Table 1, Appendix A.

Excluding the headquarters units and the Maryland Line (which on May 5 was posted in the rear along a principal line of communications with Richmond), the grand total is 52, 984 men. This return, however, does not reflect a complete listing of all the troops assigned to the Army of Northern Virginia nor all the troops that entered the Wilderness on May 5 or 6. Missing are Hoke’s Brigade, two regiments of Rodes’s Division, and more importantly all of the infantry and three battalions of artillery from the First Corps. Pickett’s Division and Read’s and Eshelman’s Battalions of artillery of this command were serving temporarily within the Department of North Carolina and Southern Virginia on April 20, along with the above-noted infantry units from the Second Corps. They were all transferred in early May to the forces assembling under Gen. P. G. T. Beauregard to defend Petersburg and Richmond against the Army of the James under Maj. Gen. Benjamin F. Butler.

The other principal omissions are Kershaw’s and Field’s Divisions and Huger’s Artillery Battalion of the First Corps. These units, along with Longstreet himself, were on April 20 rejoining the army from the previous fall and winter campaign in East Tennessee. Having reached Virginia and rejoined the main army, they were reviewed by General Lee near Gordonsville on April 29. The latest previous return available for these units in East Tennessee in the Official Records is dated March 31.¹ This return is reproduced as Table 2, Appendix A.

The sum total of these various organizations yields 63,888 men. But there is one significant problem with this number: it fails to take into consideration changes in the returns that occurred between April 20 and May 5 for the Second and Third Corps and the Cavalry Corps and between March 31 and May 5 for the First Corps. Unfortunately, the next available return for Lee’s army in the Official Records is dated June 30.² This return is reproduced as Table 3, Appendix A.

The June 30 return does not provide clear information that could assist in determining the overall army and individual unit strengths on or about May 5. The Second Corps is missing, and there are several new units that did not start the campaign in the Army of Northern Virginia. Although not of apparent immediate benefit, this later return was subsequently found to be of considerable help in determining Confederate numbers during the Overland Campaign. For this reason, it has been included.

One approach to determining the changes in Lee’s army between April 20 and May 5 is to examine the earlier returns in 1864. The available returns provided in the Official Records are dated January 20, February 20, March 20, April 10, and April 20.³ These items, and especially the last three, reveal a steady increase in the strength of the various units of the Second, Third, and Cavalry Corps during the early spring. Judging by these muster rolls, it appears that a concerted effort was made, as one would expect, to build up the manpower of Lee’s army in preparation for the anticipated 1864 spring campaign. Many of these soldiers were likely three-year veterans who had received a special furlough, a process that commenced early in the year and tapered off in the weeks leading up to May. The others were probably new personnel. It seems probable that that these units continued to experience an increase in their ranks during the two weeks between April 20 and May 5.

In order to explore and confirm this assumption, it became necessary to research and categorize the personnel who accounted for this increase in the army’s strength. This was accomplished by an extensive examination and compilation of the individual Compiled Service Records (CSRs) for all the units comprising the Army of Northern Virginia during this period. The CSRs consist of individual carded records for every soldier found on file during the war. They were compiled principally from available company muster rolls, hospital records, clothing issues, and Federal prisoner-of-war lists and are filed alphabetically by last name by unit. The full collection is currently housed at the National Archives in Washington, D.C. The cards for any individual soldier are available upon special request. More readily, though, they are available on microfilm for quick public access and study, typically presented on one or more microfilm reels for each unit.

The major component of the CSRs are the unit muster rolls. Ideally, these were filed every two months on the last day of the even-numbered months (February 28/29, April 30, June 30, and so on). The primary purpose was to determine the status of each individual soldier for pay. The availability of these rolls and the accuracy of their information vary significantly from one unit to another. For example, some of the brigades in the First Corps, Hoke’s Division (of Beauregard’s command), and the artillery have all or most of the critical rolls for 1864 on record. Other units, such as the Second and Third Corps and the cavalry, are missing one or more rolls for this year. The least-documented units were two regiments from Jones’s and Colquitt’s Brigades (from the Second Corps and Hoke’s Division, respectively). For these regiments there exists only one muster roll for the entire war, and these are from 1862. In some cases the missing rolls were lost or likely destroyed at the close of hostilities in 1865. In other cases they were clearly filed at the end of a four-or six-month period (such as May–August or May–October 1864). This is the case with most of the units in the Second Corps and Breckinridge’s Division.

Quite apart from the fact that many are missing, there are other problems with the muster rolls. In most cases the date on which the record was filed is indicated. Often this date is weeks or even months later than the period to which the return ostensibly refers. In many cases it is not clear if the recorded information reflects the status of the soldier at the end of the period listed or on the date at which the roll was filed. In other instances two or more consecutive rolls were apparently filed on or near the same date. These anomalies sometimes create confusion as to the actual status of a soldier at a specific time.

There is another problem with the information in the rolls. Soldiers are often marked “Present” even though a careful study of hospital records reveals them to be clearly disabled and unfit for field service. The actual status of these personnel can ultimately be determined with a careful study of all the available records for each soldier.

Where early rolls for 1864 are missing or clearly reflect a later period in the year, it can be difficult to determine a unit’s strength on or about May 5. One compensating record contained in the CSRs and found to be of benefit in filling in some of the gaps is the clothing issue. Clothes were typically issued for enlisted personnel on a quarterly basis. In the second quarter of 1864, this occurred in April. Soldiers for whom clothing was issued in April were likely present at the commencement of the campaign in May.

The CSRs for most of the personnel in four Mississippi infantry regiments (the 2nd, 11th, and 42nd of Davis’s Brigade and the 21st of Humphreys’s Brigade) and some of those in Law’s Alabama and Hays’s and Stafford’s Louisiana Brigades contain, in addition to the muster rolls, record-of-events cards. These were apparently prepared from memory after the war and incorporated with the other documents. They provide the status of the soldier at each of the major engagements for his unit throughout the war. Although they appear to contain some errors, they nevertheless were found to be of considerable value in determining the percentages of men present for the various battles and those sick, detached, or otherwise temporarily absent from the ranks.

Returning to an earlier theme, the individual CSRs reveal that the Army of Northern Virginia grew in strength between the last recorded return and May 5. The personnel swelling the strength of the army fell into three categories. First, there were new recruits and conscripts. The individual muster rolls for each soldier provide the date of enlistment and, in many cases for those enlisting in early 1864, the date of the actual assignment to the unit. Men returning from sick leave, recovery from wounds, or earned furlough comprised a second group. The third category was exchanged prisoners of war. An examination of their records reveals that many soldiers captured in battles in 1863 were exchanged between March 3 and 20, 1864. These men were usually sent to a Richmond hospital for examination and then furloughed for thirty days. Depending upon their furlough date and the possibility of overstaying for several days, they then returned to their respective units during the period of mid-April to mid-May 1864.

Judging by the various categories of men joining or returning to a unit, some conclusions can be reached. Most of the new enlistees to Lee’s army came from the states of Virginia, North Carolina, and South Carolina. The remaining states of the Confederacy supplied significantly fewer recruits and conscripts, their new personnel likely going to other armies in the field, such as the Army of Tennessee. Few, if any, new men joined the Army of Northern Virginia from the Trans-Mississippi states or from Federal-occupied Tennessee. In addition, the units that sustained the heaviest casualties in 1863 appear to have experienced the greatest increases in strength due to the return of many wounded and exchanged prisoners.

Despite the various shortcomings of the muster rolls and the CSRs, it has been possible to arrive at a more accurate determination of the strength of Lee’s army on or about May 5 than has previously been possible. This process involved compiling and tabulating the CSRs, comparing these results to the returns of April 20, 1864, and developing a mathematical model for adjusting the totals to reflect personnel not actually in the ranks. The best results were achieved for the Third Corps and the artillery arm. Less satisfying results were obtained with the infantry units of the First and Second Corps and the Cavalry Corps. Each of these three organizations presented specific problems. The First Corps appears to have left a trail or detached/furloughed personnel in its return from East Tennessee; judging by the CSRs, most of these men eventually caught up with their commands in May or June (during the campaign). The muster rolls for the Second Corps and the Cavalry Corps are collectively among the worst on record and lack consistency. For men of the latter command, there was the additional problem of finding a horse. Some personnel, according to the rolls, were still on furlough procuring a mount from their home area on April 30.

In applying this analytical method, the goal has been to obtain consistency. This has been achieved for the most part. Occasionally, assumptions varying from the norm were unavoidable and had to be made. These were weighed on the conservative (low) side based on the overall available data. As a further check of the accuracy of this approach, the derived strength figures were compared to the June 30 returns. Incorporating the sickness and battle losses between May 5 and June 30 together with the continuing personnel additions to the units, these derived figures were found to yield numbers that corresponded to the June 30 strengths.

In examining the individual CSRs, an obvious attempt has been made to include only those personnel who would have been in the ranks and engaged in battle. This excludes personnel who served in rear positions in the field and staff (such as quartermasters, surgeons, assistant surgeons, chaplains, ordnance sergeants, quartermaster sergeants, and commissary sergeants as well as band musicians, couriers, ambulance drivers, teamsters, headquarters guards, and others). On rare occasions, these personnel drew enemy fire and became casualties. In these limited instances, they are included with their respective company or the field and staff personnel. Using this process, the estimated strength of the Army of Northern Virginia on or about May 5, 1864, was almost 66,000 men. (See Table 4, Appendix A.) For purposes of simplicity, the unit strengths have been rounded off to the nearest 5 men. The actual individual-unit strengths (if ever found) may be around 10 to 50 men higher or lower for each infantry and cavalry unit and, similarly, around 5 to 10 men for each artillery battery.

It should be remembered, though, that the total army strength shown in Table 4 is a “present for duty” compilation and not the actual or effective combat strength. John W. Busey and David G. Martin in their work Regimental Strengths at Gettysburg (1982) indicate that for Lee’s army, the effective combat strength was about 94 percent of the “present for duty” strength for the infantry and artillery and about 88 percent for the cavalry. This reduction reflects the absence from the battlefield of stragglers and of personnel who became sick or were detached close to the start of combat. An examination of the individual records for Army of Northern Virginia personnel indicates that these assumptions are basically applicable throughout the Overland Campaign. For cavalry dismounted and employed as infantry, the effective or engaged force should be further decreased by 25 percent to account for one man in four having to hold the mounts.

There is some documented historical evidence to support this approach and to validate the mathematical model. Although it is a rather diminutive sampling, the strengths of four regiments starting out in the campaign or entering the Battle of the Wilderness are available. For example, in Louis H. Manarin and Weymouth T. Jordan Jr.’s North Carolina Troops, 1861–1865: A Roster, the 7th North Carolina Infantry Regiment is indicated as having twenty-seven officers and 425 enlisted men available for the campaign.⁴ This appears to be an approximation of the unit’s “present for duty” strength and may include some men who were detached or joined/rejoined the unit after May 5. The report of Col. James R. Hagood of the 1st South Carolina Infantry Regiment provides a second source for this study.⁵ In his report, this regimental commander states that twenty-six officers and 235 men were carried into action on May 6. The May 31 issue of the Lancaster Ledger printed an excerpt from a letter written by Chaplain H. N. Craig of the 5th South Carolina Infantry Regiment on May 9. In his correspondence, Chaplain Craig states that his regiment went into the last fight (May 6) with 270 men. A fourth source was obtained from the published diary of Joseph P. Fuller of Company B, 20th Georgia Infantry Regiment. In this document, Fuller similarly states that his unit numbered 265 men on May 6. These numbers are indicative of effective combat strengths. Divided by 94 percent, the resultant numbers compare very favorably with the “present for duty” figures for the latter three regiments.

CHAPTER 2

Reinforcements to the Army of Northern Virginia during the Overland Campaign

The Army of Northern Virginia was considerably reinforced with many new units during the Overland Campaign. Many of these were drawn from General Beauregard’s forces facing Butler’s Army of the James. Some originally belonged to Lee’s army, others were drawn from the military departments and commands along the Atlantic Seaboard. Breckinridge’s Division was temporarily loaned from the Department of Western Virginia. The identities and strengths of these units (as they joined the Army of Northern Virginia) are listed in Table 5, Appendix A. The total strength of these units was 25,495 men.

In addition to these new units, Lee’s army continued after May 5 to receive personnel to units in the original table of organization. As discussed above, the strength of the Army of Northern Virginia was increased during the immediate period preceding the commencement of the campaign by the addition both of recruits and of personnel returning from furlough and prisoner exchange. A study of the individual personnel records reveals that this process did not simply end with the start of active hostilities but continued throughout the entire year of 1864. For the cavalry arm, most of these men consisted of returnees with a new mount. For the original six brigades comprising this corps, the increases in personnel between May 5 and mid-June (after Cold Harbor/Trevilian Station) were about 120 per brigade. For Butler’s South Carolina brigade, the increase (during June only) was about 40 men. For the artillery, the increase during this period averaged about 3 to 4 men per battery. For the infantry arm, the number of new recruits and returning men varied considerably. The average increase per brigade between May 5 and mid-June (after Cold Harbor) for each infantry division is shown in Table 6, Appendix A.

Regarding the figures in Table 6, it is not surprising that the largest increases occurred in the divisions of Kershaw and Field in the First Corps. As mentioned above, these divisions left a trail of men in the wake of their movement from East Tennessee. They also had a substantial number of men returning from wounded furlough. Hoke’s Division (from Beauregard’s forces) similarly had men come in late from earlier assignments along the Atlantic Seaboard. The other larger figures are associated with commands that overall had large enrollments, sustained heavy casualties in 1863 (principally at Chancellorsville and Gettysburg), or, as with Wilcox’s and part of Heth’s Divisions, received large numbers of recruits in late May or early June 1864.

There were other minor inc...

Table of contents

• Cover
• Title Page
• Copyright Page
• Contents
• Foreword
• Preface
• Introduction
• PART ONE. Lee’s Army: An Overview
• PART TWO. Unit Discussions, with Casualty Breakdowns
• Summary and Conclusions
• Appendix A: Tables
• Appendix B: Maps
• Appendix C: Army of Northern Virginia Order of Battle during the Overland Campaign
• Notes
• Index