Civil War Tactics in Perspective


Which Civil War historians also research and write about the Napoleonic era, the eighteenth century, or other wars of the 19th century?  There are very few - but the rare ones who do give us the greatest insight into Civil War combat.  How can you understand Civil War tactics without perspective, without studying what Civil War generals studied, without comparing Civil War weapons to those that came before and after?  You cannot.  How can you understand Civil War tactics by looking solely at the infantry?  Many Civil War historians attempt just that, getting bogged down in the minutiae of battles instead of gaining perspective by researching and understanding other eras.  Because of this lack of perspective, many historians don't fully understand why Civil War combat was indecisive.  And because of their lack of background, when historians specializing in the Civil War have seen Civil War generals write of "Napoleonic" tactics, firstly - they may not have understood the sort of tactics used by Napoleon - something more than just men fighting shoulder to shoulder - and secondly, it didn't occur to them that "Napoleonic" might refer to another Napoleon, Napoleon III.  Mid nineteenth century French tactics were an evolution of earlier tactics and included a faster 'gymnastic' pace of attack to reach enemy lines faster, hopefully negating the advantages of the new rifled musket.  (The Bloody Crucible of Courage, Brent Nosworthy) 

As discussed elsewhere, the rifled musket, although a technological advance, was not the revolutionary weapon that it has been made out to be.  Artillery, on the other hand, was much improved from fifty years before.

What were Napoleonic tactics, and how do they differ from Civil War tactics?

If you understand how Napoleon fought, you will see that Civil War tactics were different in a number of important ways.  These differences perhaps explain much of why Civil War combat tended to be indecisive.

In the gunpowder age, battle was often indecisive.  Napoleonic combat, the ultimate development of linear tactics, is the exception.  Napoleonic tactics were the evolution of linear tactics born with the socket bayonet.  One hundred years before Napoleon, an infantry battalion was an unwieldy and vulnerable combination of musketeers and pikemen, with the pikemen protecting the musketeers from enemy pikemen and cavalry.  With the socket bayonet, musketeers shed themselves of pikemen - now they could protect themselves from shock attack and still fire their muskets.  Over decades, musketeers stretched themselves thinner, first into six ranks, then four, then three, and finally just two.  It was still an awkward system in some ways, with difficulty deploying an army from the march into line of battle.  In Marlborough's day, in the early 1700s, it took most of the day to prepare for battle, and it was impossible to surprise an enemy on the march.  When battle began, the whole of the infantry would typically attack together simultaneously in two lines.  The logical place for cavalry was on the flanks.  In battle, the cavalry would defeat the opposing enemy horsemen then attack the rear of the enemy infantry just like in the 1600s - but also, at least in theory, much like the American Civil War.

 Frederick the Great perfected this system, marching and deploying on the enemy's flank for a devastating attack.  Frederick's attack at Leuthen, and his later attempts to do the same, looks similar to Jackson's attack on the XI Corps at Chancellorsville and Early's attack at Cedar Creek.   Indeed, in many ways the Civil War looks more like the Seven Years War than it does the Napoleonic Wars.

 By Napoleon's time, battle was more flexible - thanks to advances in infantry tactics.

With the use of thinner battle lines - two or three ranks - a quicker, simplified system was developed to deploy from marching column into line of battle.  Battle was becoming more practical, no longer a rare and consensual event. 

Let's take a look at change in the 18th century and the development of Napoleonic tactics, one feature at a time. 


There was plenty of room to improve upon mid-18th century tactics, and the defeated and humiliated French army lead the way with reform.  The Austrians had already made great strides in reducing the weight of their cannon, with their new pieces weighing only half as much as their predecessors.  The French adopted these concepts with the Gribeauval system.  Frederick had already complaining in the Seven Years' War that the new Austrian artillery was killing off his highly trained infantry, who were renown for their discipline and quick rate of fire.  The new lighter field artillery had the potential to revolutionize warfare.  There was no choice for a monarch but to enter into an artillery arms race - or else have his army slaughtered.  Masses of the new mobile guns, grand batteries, could be brought to bear against the enemy, blowing holes in his line and demoralizing his men.


Frederick was also whining that Austrian light infantrymen sent ahead of their main line were picking off officers and men while taking cover behind trees and terrain features.  With an army made in many cases of forcibly recruited men held in bondage only by the threat of severe punishment, Frederick couldn't trust his men to go forward and skirmish.  With the prospect of freedom, they might run away and never be seen again.  Frederick resisted this change, at least for a while, before embracing it.  Other armies developed highly trained and motivated special units of light infantry too.  Although skirmishers of the French Revolutionary Wars and Napoleonic times are often stereotyped as ill-trained levies good for nothing more than swarming in dense masses in front of the enemy line, modern historians dispute this.  Whenever possible, skirmishers of the era were highly motivated, well trained, and good shots.  Their calm, slow, well-aimed fire, combined with artillery, gradually wore down the enemy and un-nerved the survivors before the main attack.


A misunderstanding of early English-speaking historians has distorted our understanding of a major aspect of Napoleonic warfare - the use of columns.  Focusing on the Peninsula War and relying almost exclusively on British sources, some historians came to the misguided conclusion that the French primarily used battalions of infantry for shock action in columns of attack.  As early as the mid 1700s the French were keeping their second line in column formation to allow for a flexible response to problems or opportunities on the first line.  During the 1777 Battle of Brandywine, the British army, generally thought of as conservative, advanced on the rebels in maneuver columns then deployed into an open-order line.  By Napoleonic times French infantry, preceded by skirmishers, advanced toward the enemy in battalion-sized columns for more flexibility and maneuverability on the battlefield, allowing them to adjust to enemy deployments, better seize opportunities, or attempt to flank the enemy line.  Before contact with the enemy, however, the columns would deploy into line for the firefight.  A bayonet charge in column formation was only done after gaining firepower superiority when the enemy troops looked shaky and on the verge of breaking.  Against most armies, this method worked very well.  The infantry's increased flexibility helped in other ways.  The infantry columns also allowed much closer cooperation with the artillery, which was better able to mass against a portion of the enemy line or even move forward with the infantry.  Closer co-operation with the cavalry was also possible.  Unlike in the 18th century, cavalry could advance right along with the infantry, and the infantry was not restricted to an attack by the whole army. 


While 18th century cavalry was placed exclusively on the flanks, the new infantry columns allowed the mounted arm to advance in columns directly in support of the infantry, and even change places with them to lead the attack.  To protect themselves. enemy infantry would form squares, but in so doing they deprived themselves of much of their firepower and their ability to maneuver.  If artillery could be brought forward, the enemy squares would be blasted out of existence.  Either way, a massive hole was formed in the enemy line, and decisive victory was assured.

In terms of grand tactics, Napoleon would typically threaten the enemy's flank, forcing him to commit his reserve.  He always kept a large reserve available, a vitally important part of his system, to either exploit success or stave off defeat.  Most often, with the enemy reserve committed, Napoleon would send his own reserve into the weak point in the enemy line and secure a decisive victory.  Co-operation among all three combat arms was key to Napoleon's system, and the reserve decided the battle.  Does this sound like Civil War tactics to you?  No, far from it!  Civil War armies kept few reserves, and Civil War combat featured little in the way of combined arms cooperation.  Unlike the Napoleonic Wars, and more like the 18th century, reserves were a rarity in the Civil War, and a commander had few options once a battle "developed" to maturity.  Civil War tactics were NOT Napoleonic, at least not in the sense of Napoleon I.

Map of the Battle of Waterloo, 1815

Note that Napoleon's army (in blue) has cavalry not only on the flanks but also behind the infantry.  The same is true for Wellington's army.  Also note that Napoleon has a large reserve composed of Lobau's corps and the Imperial Guard.  This deployment is very unlike that of a Civil War army.


Map of Bussaco, 1810 - Napoleonic Tactics Gone Bad

In the rough terrain of the Iberian peninsula, French artillery was less effective, and cavalry couldn't cooperate very easily with infantry.  Here at Bussaco, large numbers of British skirmishers hid the infantry line and allowed a surprise volley followed by a decisive bayonet charge on the un-deployed French columns.


French tactics in Spain and Portugal didn't work out so well, and their old school British opponents may have never understood their intent.  Rough terrain complicated the system considerably.  It was difficult to mass artillery batteries, and it was difficult for cavalry to support the infantry.  Wellington sometimes sent as much as one third of his men forward as skirmishers, and the French could never gain superiority over them.  Since Wellington kept his infantry hidden behind the reverse slope, the French maneuver columns blundered upon the British line and were surprised.  Struggling to deploy into line and disorganized and shocked by the surprise encounter, the French received one or two close range volleys from their British opponents who then let out a cheer and charged.  The British system worked consistently, except at Albeura - and at Waterloo, where the open terrain suited Napoleon's methods.  Terrain explains much of the British success against the French.  In part, terrain also explains why combined arms tactics were so rare during the Civil War, but we'll discuss that later.  Perhaps because of the British success, and because of the bloody nature of Napoleon's later battles, post-Napoleonic military theorists became somewhat reactionary and supported tactics which looked backward to the 18th century.  Napoleon's later battles became less decisive, partly because his enemies adopted his methods, and partly because the greater firepower from more artillery made attacks less likely to succeed.



Civil War Tactics Look Like a Poor Man's 18th Century


The first American army, the Continental Army, was based on its British opponent during the Revolution; even then, the British were known for a brief musketry exchange followed by a charge.  Civil War tactics don't resemble this, or British Peninsula tactics, or Napoleonic tactics.  But Civil War tactics do resemble the 18th century in that infantry was typically formed in two lines flanked by cavalry.  Civil War battles sometimes even featured a Frederican style oblique order attack.

Although Gettysburg featured a Napoleonic concentration of artillery, genuine Napoleonic tactics for example, would have involved cavalry supporting Pickett's Charge - and Union cavalry charging in pursuit of the defeated Confederates- in short, it would have been combat with the potential for decisiveness.  The technology of rifled muskets didn't make this sort of thing impossible.  Koniggratz in 1866 shows this as do some later Civil War battles like 3rd Winchester.  Because combined arms attacks weren't regularly attempted, battle was predictably indecisive.

Unlike Napoleonic and eighteenth century battle, all too often, Civil War combat degraded into a confused infantry firefight with officers gradually losing control, with any hope for maneuver lost.  After the onset of confusion, shock action with the bayonet was not practical.  Due to lack of training and discipline - an inevitability to some degree with a volunteer army of a democratic government - it was always difficult to get men to close with the enemy.  Once an infantry advance stopped in order to fire, it could rarely be made to continue forward.  GFR Henderson wrote on page 215,

"Occasionally, when protected by unusually strong defenses, the leaders were able to induce their men to reserve their fire to close range, but, as a general rule, whether defending or attacking, the men used their rifles at will.  The officers were never sufficiently masters of their soldiers to prevent them, when bullets were whistling past, from immediately answering the enemy's fire.  In the best Confederate regiment, in the midst of a conflict, the ardent and burning inclination of the soldiers was obeyed rather than the commands of the officers."

Prussian observer Justus Scheibert believed that a deficiency in lower level officers, who showed "ignorance of military things", explained why the brigade became the tactical unit of the war, "hence stiffness in the lines and clumsiness in management and direction of troops".  Poor performance on the battlefield was the result, and "the loss of an upper-level commander would cripple (the) advance".  He described an attacking infantry division as "like ghosts of days and ways of Frederick the Great.", in essence a poor man's version of mid 18th century methods.  (Scheibert 49)  He described a typical attack;

 "The nearer to the enemy, the more faulty the lines and the more ragged the first (line) until it crumbled and mixed with the skirmishers.  Forward went this muddle leading the wavy rest.  Finally the mass obtruded upon the point of attack.  In a sustained, stubborn clash, even the third (line) would join the melee.  Meanwhile the usually weak reserve tried to be useful on the flanks, or stiffened places that faltered, or plugged holes.  In sum it had been a division neatly drawn up.  Now its units, anything but neat, vaguely coherent, resembled a swarm of skirmishers." (Scheibert 41)

In contrast, Scheibert writes:

"Prussian tactics freed (officers) to use their own minds...  Liberated battalion and even company commanders could be the heads of tactical units, their own, and make them fight as right-thinking officers saw fit and as well-trained troops best could.  The flexible line at the forward edge resembles a chain, then with detachable links under independent guidance.  At crisis they can dismember into smaller and even the smallest units without dysfunction...  Our Prussian tactics thus gave our line officers energy, elasticity, and speed - to the entire army's benefit...  Furthermore, diligent peacetime training provided our troops an abundance of formations, something to fit any circumstance...  Lee, the first American to acknowledge this superiority, replied in the thick of Chancellorsville when I spoke with amazement at the bravery of Jackson's corps: 'Just give me Prussian formations and Prussian discipline along with it - you'd see things turn out differently here!" (Scheibert 49)  

An extreme example of how potentially decisive combat degraded into chaos is Brawner Farm.  Jackson had the opportunity to attack and crush an isolated and much smaller Union division with his corps, but the attack stalled, and an indecisive firefight resulted, and because of Jackson's peculiarities, his subordinates feared to take the initiative and stood idly by while the opportunity to destroy a Union division was lost.

The human element is important in combat.  Men are not machines, and American volunteers were not European professionals.  The Comte de Paris wrote in his "Campagne du Potomac" on pages 144-4:

"The will of the individual, capricious as popular majorities, plays far too large a part.  The leader is obliged to turn round to see if he is being followed; he has not the assurance that his subordinates are bound to him by ties of discipline and of duty.  Hence come hesitation and conditions unfavorable to daring enterprise."

 On a more personal level, a New Yorker in "Battles and Leaders, vol 2, p662" wrote:

"The truth is, when bullets are whacking against tree trunks and solid shot are cracking skulls like egg shells, the consuming passion in the heart of the average man is to get out of the way.  Between the physical fear of going forward, and the moral fear of turning back, there is a predicament of exceptional awkwardness, from which a hidden hole in the ground would be a wonderfully welcome outlet."

What could negate these natural tendencies?  GFR Henderson explains, "Mutual confidence is the force that drives a charge home; and this quality is the fruit of discipline alone, for in almost every campaign it is the better-disciplined troops who have displayed the greatest vigor in assault."  Henderson also cites (on p216) a comment by Lord Wolseley, who stated that a single corps of regulars on either the Union or Confederate side, would have won the war on account of their superior mobility and cohesion - both traits coming from discipline.  In James I Robertson's "Soldiers Blue and Gray", page 123, early in the war Joe Johnston is quoted, "I would not give one company of regulars for a whole regiment!"  Relying on an army of volunteers enlisted at the start of wars was perhaps the country's only choice, but the quality of Civil War armies was questionable, and this was not a new thing.  In "Civil War Infantry Tactics", Earl Hess points out that the army of the War of 1812 did poorly with the sole exception of Jacob Brown's army near Niagara, and in the Mexican War, "The volunteers gained a well-deserved reputation for lacking discipline, marauding, committing atrocities against helpless Mexicans, and running away from battlefield dangers."    

Could Civil War battles have been decisive?


Maybe.  Skirmishers weren't consistently well used.  In the French Revolutionary Wars and the Napoleonic Wars, skirmishers could all but decide a battle, but we don't see that in the Civil War.  The Confederate army improved in this respect through the war, forming elite sharpshooter battalions in each brigade, but the Union army allowed its light infantry to decline over time.  Why couldn't the Union army have formed units similar to the Confederate brigades' sharpshooter battalions and one-upped them by giving their skirmishers more repeating rifles?  Properly organized Union skirmishers with repeaters firing rapidly from a prone position might have easily dominated Southern light infantrymen - and possibly even repulse full scale attacks.  Could the cavalry have advanced with and supported infantry like in Napoleon's time?  If you accept the argument that rifled muskets were little better than smoothbores, then maybe they could.  Although admittedly an occurrence of the smoothbore era, Jeb Stuart successfully attacked Union infantry at 1st Manassas.  Later examples were not as successful.  At Gaines Mill and at Cedar Mountain, Union cavalry unsuccessfully attacked infantry, but these failures should come as no surprise since these were desperate attacks by small units against unbroken advancing infantry.  British observer Arthur Fremantle in "Three Months in the Southern States" on page 157 recounts meeting infantry in the Western Theater,

I expressed a desire to see them form square, but it appeared they were "not drilled to such a manoeuvre" (except square two deep).  They said the country did not admit of cavalry charges, even if the Yankee cavalry had stomach to attempt it. 

Fremantle (p 159) met a Western Theater Confederate cavalry colonel:

He explained to me the method of fighting adopted by the Western cavalry, which he said was admirably adapted for this country; but he denied that they could, under any circumstances, stand a fair charge of regular cavalry in the open.

Then the colonel goes into explaining dismounted fighting and bluffing the enemy.

Fremantle moved to the Eastern Theater and visited Stuart's cavalry.  On page 256-7 and 292:

I remarked that it would be a good thing for them on this occasion they had cavalry to follow up the broken infantry in the event of their succeeding in beating them.  But to my surprise they all spoke of their cavalry as not efficient for that purpose.  In fact, Stuart's men, though excellent at making raids, capturing waggons and stores, and cutting off communications, seem to have no idea of charging infantry under any circumstances...  The infantry and artillery of this army don't seem to respect the cavalry very much, and often jeer at them...  Staurt's cavalry can hardly be called cavalry in the European sense of the word; but, on the other hand, the country is not adapted for cavalry.

Perhaps fences and broken terrain with numerous woodlots made cavalry attacks on infantry impractical.  At Chancellorsville, for example, a single Union cavalry regiment attacked down a narrow road, unable to change direction.  Failure was the predictable result.  Lack of training and experience in the cavalry arm may also have made it impractical.  In the Confederate service, the men owned their horses, making them less willing to risk them.  Regardless of who paid for them, horses were expensive, and it was difficult to find forage for them in a country more sparsely populated than Europe.  European cavalry was the product of years of training, a horse alone requiring three years to train.  Further, Union cavalry were armed with repeating rifles by late war, and in many cases served as mounted infantry, which meant that they largely abandoned the full potential of shock attack, their traditional battlefield role.  Late in the war, the Union cavalry did occasionally mount large scale shock attacks against infantry and conducted after-battle pursuit - at Third Winchester, Cedar Creek, and Sailors Creek to great effect, for example.  We may never know what could have been.  

Union cavalry was handicapped from the very beginning of the war.  In no small part the problem was Winfield's Scott's attitudes toward cavalry.  Believing that cavalry was obsolete because of advances in weapons technology, Scott raised few cavalry regiments at the beginning of the war - a war that would soon be over, so he thought.  "Kearny's Dragoons Out West" by Will and John Gorenfeld discusses the raising of a dragoon regiment in the 1830s.  There was resistance to forming a dragoon regiment because cavalry was perceived as expensive, European, and aristocratic - everything that America claimed not to be. (p 23)  Minimum training time was twenty-three weeks.  (p294)  The Mexican War saw the 1st Dragoon Regiment broken up, never to fight as a whole regiment.  Five companies went to New Mexico, two to Taylor's army, and one company initially assigned to Taylor going to Winfield Scott's army.  (Gorenfeld p304)  Only with the rise of McClellan were significant numbers of cavalry regiments raised.  Even then, it took time to train them, and many were used to protect lines of supply or were dispersed throughout the army. 

It was only late in the war that the Union army had large bodies of cavalry available.  Even then, it wasn't always properly used.  Having a good body of cavalry available for the Overland Campaign, Grant sent it on a raid to Richmond, losing not only any usefulness on the battlefield but also its usefulness with screening and reconnaissance.  Then, facing Lee at Cold Harbor, he once again sent his cavalry on a costly raid that was repulsed at Trevalian's Station.  The French observer DeChanal in "The American Army in the War of Secession" was skeptical of raiding. 

"After Sheridan's raid in Virginia, an expedition which lasted more than a month, all the unserviceable horses and broken down horses were gathered together in a park at City Point; there were quite six thousand of them." (p29)

"The Quartermaster's Department is charged with the duty of supplying horses for the cavalry.  The number required surpasses all belief, but it is partly explained by the raids.  These are expeditions in which the cavalry lives on the country, traveling many miles in a few days, sometimes without finding water or suitable forage.  The greatest reason for this enormous waste of horses, is the lack of intelligent care, natural to an inexperienced horseman." (p 30-31)

 DeChanal goes on to point out that between May and October of 1863, the effective strength of the cavalry of the Army of the Potomac fluctuated between 10,000 and 14,000 but the number of remounts required was over 35,000! 

"This is the equivalent to a loss of two and one-half horses per man, or a rate of five horses per annum.  And in addition, it must be remembered that this does not include horses captured from the enemy, which at its average, would probably amount to the entire remount for two months." (p31)

What of the raids themselves?  DeChanal cites a raid in which on third of the horses died of thirst.  In another,

"General Wilson was sent to the left of the lines of Petersburg to cut the railroads.  He succeeded in this but was surrounded by a large force of infantry, and was able to escape with only 3,000 cavalry, scarcely half his force, abandoning his artillery, his prisoners, and 2,000 unfortunate negro refugees who were driven back with strokes of sabre and lash to Richmond and sold at auction." (p 235)

Were the raids worth the cost?  For a "weaker enemy", the Confederates, for example, raids to cut communications and destroy supplies were useful.

"But in the midst of serious operations, to detach a considerable force to make a diversion which usually is not an appreciable factor in the final result, is an operation of doubtful utility.  Had Sheridan's force been with Grant at Spottsylvania, a decisive result might have resulted.  Moreover, the damage resulting from the destruction of railways is often more apparent than real.  After Wilson's disaster, the Federal forces felt consoled by the fact that his expedition had destroyed the Danville railroad; eight days later it was again in operation." (p 235) 

Which is to say that no only was cavalry not used for combined arms action or effectively for pursuit, its usefulness for screening and reconnaissance was often discarded in favor of raids of questionable value.

C-Cubed and Staffs


Earl Hess in "Civil War Infantry Tactics" points out the fact that Civil War armies had great difficulty coordinating large attacks, having trouble coordinating divisions and corps - much less whole armies.  In the West, the Confederate Army of Tennessee had trouble coordinating attacks above the brigade level while its Union counterpart, although better, nevertheless gained victory operationally rather than tactically.  In the East, the Confederate army coordinated attacks at the brigade level at Seven Pines and Malvern Hill and at the division level at Second Manassas and Pickett's Charge, but in 1864 and 65, counterattacks were typically at the brigade level - a regression rather than a progression.  The Army of the Potomac, Hess argues, was the only army of the war to regularly show competence in coordinating attacks.  At Five Forks, for example, Gen Warren distributed copies of a map showing the plan of attack; even still, Warren lost control of two of his three divisions and was relieved of command by Sheridan as a result - for remaining in place so his subordinates would be able to find him.  Difficulty coordinating - "articulation" is the word that Hess uses, made flank attacks and the exploitation of success difficult.  Frustratingly, despite making these arguments, Hess then downplays the consequences of the failure of Civil War armies to "regularly organize large formations on the corps level" and makes no attempt to explain the reasons for this failure.  What were the reasons?

Issues of staff, command difficulties, military education, and philosophy of command are key to understanding this; they were as important as technology and tactics - perhaps more so.  A general's staff was like the nervous system of the army; an inadequate staff made it virtually impossible to control an army.  Sharing the same background, commanders on both sides fought using staffs that were much smaller and much less competent than their European counterparts.  For a more lengthy treatment, please see "Staff and Headquarters in the Civil War".



The Civil War was not particularly advanced tactically, but it was not fought using the tactics of Napoleon I either.  Regardless of the reasons - technology, tactics, terrain, command and control problems - or more likely a mixture of all these things - circumstances tended toward making Civil War combat less decisive than Napoleonic combat.  Whether Napoleonic combined arms cooperation was still possible on a large scale during the Civil War is debatable.  But the advent of rifled muskets doesn't explain this failure.  Advances in artillery technology had more effect, but even these changes don't offer a complete explanation.  Command, control, and communications problems, along with difficult terrain, made decisive battle more difficult to achieve.  When victory was achieved, pursuit was often half-hearted and limited to the speed of infantry since cavalry was scarce.  Numerous are the examples of Civil War battles on the verge of decisiveness - but without that final step that would have annihilated the enemy - Shiloh, Second Manassas, Antietam, Gettysburg, Chickamauga.  Perhaps this failure to achieve decisive results encouraged the late-war custom of entrenching, although there are other explanations.  The Prussian observer Justus Scheibert argued that breastworks were "safeguards against panic". (Scheibert 50)  Earl Hess argues that troops entrenched as a reaction to the shock of battle.  The effectiveness of artillery perhaps contributed to the practice.  Others point out that entrenching freed up troops to turn or flank the enemy - so entrenching was essentially a method to facilitate maneuver - one which had the opposite effect.  GFR Henderson saw a relationship between entrenching and enemy discipline.  He wrote, 

"Very early in the War of Succession, the Federal commanders, recognizing their enemy's disposition to bring matters to a speedy issue, made use of earthworks and entrenchments; the Confederates, at a later period, when the desperate assaults on the Fredericksburg heights taught them that the Northern battalions had at length learnt to follow their officers to certain death, gave up their trust in broken ground and sheltering coverts, and adopted the same means of stiffening the defence.  In 1863, the third year of the war, both armies became equally formidable on the defensive, ... (and) the confusion of the earlier fields of battle was no longer seen."

Only one side had to entrench in order to force their opponent to do so.  To do otherwise was just too risky, and there was no turning back.  All of these explanations above have validity.  The 1864 campaigns little resemble those of 1862 or 1863.  Battle lines were stretched thinner, putting commanders even more out of touch with the situation, and making armies even more difficult to control than before.  At any rate, battle tactics had failed.  Perhaps the use of entrenchments was inevitable, with a bloody attritional struggle like the Overland and Petersburg Campaigns being the predictable result.  Eventual Union victory was the result of operational successes like Vicksburg and Appomattox - and not from the destruction of enemy armies on the battlefield. 




Sources and Suggested Reading:

Michael A. Bonura, Under the Shadow of Napoleon

Bowden and Ward,  Last Chance For Victory

David Chandler, Art of Warfare in the Age of Marlborough, The Campaigns of Napoleon

Phillip Cole,  Civil War Artillery at Gettysburg, Command and Communications Frictions in the Gettysburg Campaign

Jean Colin,  Transformations of Warfare

DeChanal, The American Army in the War of Secession

Christopher Duffy,  Instrument of War: The Austrian Army in the Seven Years War

Robert P Epstein, Napoleon's Last Victory and the Emergence of Modern War

Lee W. Eysturlid, The Formative Influences, Theories, and Campaigns of the Archduke Carl of Austria

Steven Fratt, The Guns of Gettysburg - North & South August 2004

Arthur Fremantle, Three Months in the Southern States

Gates, David, The British Light Infantry Arm, c. 1790-1815

Gorenfeld, Will and John, Kearny's Dragoons Out West

Paddy Griffith  Battle Tactics of the Civil War, Forward Into Battle, Battle

Edward Hagerman, The Civil War and the Origin of Modern Warfare

William Hazen, A Narrative of Military Service

GFR Henderson, The Science of War

Earl Hess, Field Armies and Fortifications in the Civil War

Earl Hess, Civil War nfantry Tactics

Ian Hope, A Scientific Way of War

Wayne Hsieh, West Pointers and the Civil War

BP Hughes,  Firepower

Robert K Krick, Stonewall Jackson at Cedar Mountain

Brent Nosworthy Anatomy of Victory, With Cannon Musket and Sword, The Bloody Crucible of Courage

Peter Paret, The Cognitive Challenge of War: Prussia 1806

Christopher Perello, The Quest for Annihilation

Robert Quimby, Background of Napoleonic Warfare

Fred Ray, Shock Troops of the Confederacy  

Carol Reardon, With a Sword in One Hand and Jomini in the Other

Justus Scheibert, A Prussian Observes the American Civil War

Moxley Sorrel, Recollections of a Confederate Staff Officer

Arthur Wagner, Organisation and Tactics

Geoffrey Wawro,  The Austro-Prussian War, The Franco-Prussian War



Copyright 2008-24, John Hamill


Back to Civil War Virtual Battlefield Tours