The ground between Seminary Ridge and Cemetery Ridge was roughly a mile of open farmland. On the morning of July 3, 1863, thousands of Confederate soldiers lined up along the treeline, their bayonets glinting in the sun. Across that field, the Union army waited behind stone walls and emplaced artillery. The attack that would become known as Pickett’s Charge was about to begin.
One man believed he already knew exactly how it would end.
Lt. Gen. James Longstreet looked across that field and saw a slaughter before a single soldier stepped forward. He had told his commander, Robert E. Lee, as much in three separate conversations over three different days. He was not guessing. Twice before in this war, he had stood in the middle of that same kind of ground. Once on the winning side, once on the losing side. He told Lee that 15,000 men could not take that position. He was not offering an opinion. He was reporting an outcome.
Lee listened each time. And each time, he sent the men forward anyway.
This is not a story about a warning that went unheard. It is the story of a man who told the one person capable of stopping the attack exactly what would happen, and was overruled by a commander who heard every word.
Longstreet was not an ordinary general. Lee called him his “old war horse.” Born in South Carolina in 1821 and raised on a Georgia farm, Longstreet was tall, broad-shouldered, and quiet. He let his record speak. He graduated from West Point in 1842, relying more on leadership than academics. At the Battle of Chapultepec during the Mexican-American War, he charged through enemy fire and came out limping with a reputation for courage.
When the war broke out in 1861, he resigned his U.S. Army commission and rose quickly through Confederate ranks. His corps helped save Richmond during the Seven Days Battles. His counterattack broke Union lines at Second Manassas. He held under immense pressure at Antietam.
Then came December 13, 1862, at Fredericksburg, Virginia.
Longstreet’s men held a sunken road at the base of a rise called Marye’s Heights. They packed into ranks behind a low stone wall. In front of them stretched 400 yards of open ground. Artillery was positioned above. Before the battle, Longstreet’s artillery chief, a colonel barely past 20, told him the ground was so completely covered that a chicken could not survive it. That is Edward Porter Alexander’s later account of what he said. No one wrote it down that afternoon.
What happened next is not disputed. Wave after wave of Union soldiers crossed that field through the afternoon. Not one reached the wall. By dusk, the Federals had suffered nearly 13,000 casualties, most of them on that slope. Lee, watching nearby, said the line everyone remembers: “It is well that war is so terrible, or we should grow too fond of it.”
Longstreet spent that evening on the winning side of the exact thing Lee would one day order him to attempt. Remember the name of that young colonel, Edward Porter Alexander. He would reappear.
There was a deeper knowledge beneath Longstreet’s professional experience. In January 1862, scarlet fever swept through Richmond. Longstreet’s wife Louise was in the city with their four children. All four fell ill. Mary Anne died first, on January 24. She was 13 months old. James died the next day. He was four. Gus, Augustus, was 11. He held on for another week and died on February 1. Only Garland, age 13, survived the illness.
The bodies were placed in a borrowed vault at Hollywood Cemetery. The kind of arrangement you make when you cannot manage anything more permanent. It was supposed to be temporary. It lasted eight years.
The Longstreet who returned to the army that February was not the same man who had left. Before Richmond, his headquarters had been the liveliest in camp. Poker every night, whiskey, officers laughing late into the evening. After, he stopped drinking and turned to the church. He became quiet. The quiet never really left him.
None of this is speculation about what was in his mind. It is the documented shape of a man’s life after burying three children in eight days. What it points to is more specific. A man who had already learned in Richmond how heavy casualty estimates become when you are the one who has to carry them.
His doctrine was summed up in a sentence he wrote himself. Offensive strategy, defensive tactics. Carry the war to the enemy, then make the enemy come to you. That was the application of Marye’s Heights. At Gettysburg, that doctrine would collide head-on with Lee’s. This time, being right would not be enough to be heard.
After the costly victory at Chancellorsville in May 1863, which cost Stonewall Jackson his life, Lee proposed something bold. Move the war out of devastated Virginia, feed the army from Pennsylvania farms, and perhaps force Washington to negotiate by threatening it directly. Longstreet disagreed. He wanted to send troops west instead, to reinforce Bragg in Tennessee or break the siege around Vicksburg. Lee rejected the proposal.
Longstreet agreed to the invasion on one condition, at least according to his later account. That the campaign remain defensive. That Lee had promised to maneuver the Federal forces into attacking a position of the Confederates’ choosing. Lee would later deny saying this, and no written order survives to prove which man remembered correctly.
What the two men agreed on needed no promise. Both knew the army was marching into enemy territory with no support behind it, no supply line, no railroad. Every bit of food came from farms they would eventually have to abandon. Lee looked at that and concluded he must force a decision while forcing one was still possible. Longstreet looked at it and concluded the opposite. An army in that condition could not sacrifice men on ground the enemy had chosen, because it had no way to replace them. The argument they would have at Gettysburg had already been settled before either man saw the town.
They marched north blind. Jeb Stuart had taken the cavalry on a raid, and without it, Lee had no way to screen the Army of the Potomac. They also marched north under orders Lee intended to be taken seriously. Guards were posted at prominent houses. Soldiers caught entering towns without permission faced disciplinary action. Supply officers paid for what they took in Confederate currency, worthless as it was, with signed receipts rather than outright seizure. Lee understood that the invading army’s behavior was itself a weapon. If the Confederacy hoped for European recognition, or for Northern peace advocates to see this as a legitimate war rather than a raid, his soldiers had to act like an army, not a mob.
Through towns like Chambersburg, where civilians openly mocked the passing columns, that discipline mostly held. It held toward white civilians. The same army issuing signed receipts to Pennsylvania farmers was, in those same weeks, kidnapping free Black residents of Pennsylvania. Men, women, and children who had never been enslaved, or who had escaped and built lives in the North, were sent south into bondage. This was not random misconduct winked at by officers. It was organized, and officers did the organizing. Free people were treated as fugitives and marched away as property. This does not appear in most accounts of the campaign, including this one for a long time. That is not an accident, and it is worth keeping in mind.
The battle began by accident. A.P. Hill’s men collided with Union cavalry west of Gettysburg on July 1. What started as a skirmish became a full engagement. Union General John Reynolds was killed rushing up reinforcements. By evening, Ewell’s corps had pushed the Federals back through town toward the heights behind it, Cemetery Hill and Culp’s Hill. Lee arrived that evening and told Ewell to take the hills if possible. Ewell hesitated. He did not take them. That decision would shape everything that followed.
Meanwhile, Meade, the new Union commander who had been in post only three days, fortified his army in a fishhook-shaped line running from Culp’s Hill to Cemetery Heights and on to the Round Tops south of town. Longstreet reached the field that evening, ahead of his troops, and watched the Federals settle onto those heights. He had built a similar position himself seven months earlier, using a sunken road and a stone wall.
He delivered his first warning to Lee. Move around the Union left flank. Stand between Meade and Washington. Force Meade to attack on ground the Confederates chose. “All we have to do is move around his left and secure good ground,” Longstreet said, at least according to his 1896 memoir, published three decades later, a point at which he had every reason to want history to remember that sentence. Lee heard him and disagreed. “If the enemy is there tomorrow, we must attack him.”
Longstreet pressed again. If the enemy is there, it is because he is anxious for us to attack him. That is a good reason, in my judgment, not to do so. Lee did not yield. He simply moved on. The first warning, heard and ignored.
By the morning of July 2, Longstreet’s divisions under Hood and McLaws had arrived, exhausted from a forced march. Lee’s plan called for successive attacks. Longstreet would hit the Union left, Ewell would hit the right at Culp’s Hill. Longstreet made his argument a second time, tracing a route around the Round Tops that would avoid Union artillery entirely. Lee’s finger landed on a different line, straight up the Emmitsburg Road into the heart of the Union position. “Attack along the Emmitsburg Road,” he said. Final decision.
Longstreet set about executing an order he did not believe in. According to his subordinates’ later accounts, the tension showed. Hood remembered him saying, “As I rode beside him before the attack, the general is a little nervous this morning. He wants me to attack. I do not want to do so without Pickett. I never like to go into battle with one boot off.” Hood wrote that in his own 1880 memoir, 18 years after the event, with his own reasons for wanting history read a certain way. Take it as Hood’s memory of the moment, not a transcript, but it is consistent with everything Longstreet said about that morning closer to the battle itself.
Pickett’s division was still on the road, the last of Longstreet’s three divisions, 4,500 men who had not yet fired a shot in Pennsylvania. Longstreet had been asking for it since the day before. He would keep asking until he got it. He should have been more careful what he wished for.
The fighting that evening gave Longstreet the last piece of what he thought he needed. Hood’s men drove hard into the Devil’s Den and the Wheatfield. Bayonets, rifle butts, and rocks were used when nothing else remained. Hood himself fell early, his arm shattered. McLaws slammed into the Peach Orchard, where Union General Dan Sickles had pushed his corps forward without orders and left it exposed in the open. They shoved the Union line back violently. Then they stopped. Little Round Top held, where Colonel Joshua Chamberlain and the 20th Maine met them with bayonets after running out of ammunition. On the Union right, Ewell’s attack gained ground on Culp’s Hill and Cemetery Hill but stalled in the dark.
Longstreet rode along that line under fire, steadying his men through the worst of the battle. He later described it as the best three hours of fighting he ever saw any troops do on any field. Still, the attack failed to break the line. A full-strength striking force hitting a weaker part of the Union position on ground that offered some cover, and the line bent without breaking. That was the evidence in front of him. His own casualty reports against a position that had not moved.
He did not go to Lee that night, and according to his often-repeated later account, he received no orders either. A fact he spent the rest of his life clarifying to those who accused him of ignoring an order issued at dawn, orders for which no surviving paper proves were ever issued. What he did that night is the point where records become less clear than popular accounts suggest. Longstreet later insisted he sent horsemen south in the dark to find a route around the Union left, and that by morning he still believed or hoped that maneuver was possible. No maps, orders, or staff records survive to confirm exactly what those scouts found, or how far south they went, or whether a usable road existed behind the Union line at that hour on that ground with Union cavalry guarding the flank.
What is certain remains more limited. This was the alternative Longstreet kept proposing on July 1, again that night, and one final time the next morning. Lee kept refusing it.
By dawn on July 3, Lee had different news anyway. Ewell had lost the ground gained on Culp’s Hill overnight. Everything left of the second day’s plan had evaporated. Longstreet spoke first. He told Lee his scouts had been out all night, and he believed there was still a way to swing around the Union left and force Meade out to attack them instead. Lee raised his hand and pointed northeast toward the hill. “The enemy is there, and I am going to strike him,” he said. Everything would center on the middle.
Longstreet said, “General, I have been a soldier all my life. I have been with soldiers engaged in fights by couples, by squads, companies, regiments, divisions, and armies, and should know as well as anyone what soldiers can do.” His 1896 memoir records the conclusion he reached, that the 15,000 men who could make a successful attack over that field had never been collected for battle. This is one of the most documented complaints in the entire story, not only because Longstreet repeated it in his official writings. A British officer traveling with his staff, Arthur Fremantle, recorded Longstreet making nearly the same argument aloud the night before, that the mistake was attacking with 15,000 men on ground that required 30,000. Fremantle’s diary was published before the war even ended. Two accounts, years apart, from two different men, agreeing on the same number.
Longstreet warned that pulling Hood and McLaws off the flank to support the attack would leave the Confederate left open to 20,000 Union troops who could crush the column from the side. He said if this was to happen at all, it would take 30,000 men, not 12,000. Lee heard the number and waved it away. “They will break any line,” he said.
Three warnings, heard three times, ignored three times.
Longstreet would command the attack. The division at the center of it, the fresh division he had spent two days requesting, would be Pickett’s. Brigades under Pettigrew and Trimble from Hill’s corps would complete the formation. About 12,500 men behind the largest artillery bombardment the army could muster.
As Lee presented this plan, Longstreet recognized it because he had helped design it once before. A year and two days earlier, on July 1, 1862, at a place called Malvern Hill outside Richmond. Lee had sent Longstreet to reconnoiter the Union position on the right while Lee took the left. They met back at headquarters at noon to settle the plan, and both returned saying the same thing. Those Federal guns had to be smashed by Confederate artillery before any infantry approached. So that was the plan. Two large batteries positioned to hunt Union guns in a crossfire. Bombard until the enemy artillery was broken. Then send the infantry up the slope with Lewis Armistead’s brigade carrying the signal.
Every part of that plan failed. The Confederate guns were never massed. Most never reached the field, and those that did arrived piecemeal and were crushed. One of Lee’s generals described the performance as laughable. The signal was given, but half the brigade commanders never received it. The infantry went up the slope anyway, toward guns that had not been touched. About 5,600 Confederate soldiers fell that afternoon. Another general who walked that field later said it plainly. It was not war. It was murder.
The Union artillery at Malvern Hill had been arranged by a Federal officer named Henry Hunt. Hunt was now Meade’s chief of artillery on Cemetery Ridge, directly in front of where the Confederate infantry would head in a matter of hours. Longstreet had direct experience with what happens when that specific plan fails. Lee had studied the same disaster and drawn a different lesson. The orders had been wrong. The guns had stayed silent. Fix the coordination, mass the guns properly this time, and the result would change. Two men watched the same failure seven years earlier and learned different things. That, as much as anything, is the argument that decided July 3.
“I never felt so depressed as I did that day,” Longstreet told his staff later. Duty-bound, he tried to put one safeguard in place. He sent word to his artillery chief, Colonel Edward Porter Alexander, the same officer who had promised him at Fredericksburg that a chicken could not survive the open field. That note still survives in Alexander’s papers, in Longstreet’s handwriting. One of the few moments in this story where no one has to trust anyone’s later account.
“If the artillery fire does not disable the Union guns,” it read, “do not advise Pickett to make the charge.” Longstreet went further. He said he would rely heavily on Alexander’s judgment to decide the matter and expected him to notify Pickett when the moment came. The corps commander had just handed the launch of the largest attack of the war to a colonel under 30, asking him to make that judgment nearly blind, reading entrenched Union lines through the smoke of cannon fire that would obscure most of the field.
Alexander understood the weight immediately and objected in writing. He wrote that if there was any alternative to this attack, it should be studied carefully before a single gun fired. Longstreet did not withdraw the note. Later, standing beside Alexander as the guns worked, he said the rest aloud. He did not want to make this attack. He did not think it could succeed. He would not make it except that General Lee ordered it and expected it. Alexander, according to his account written 14 years later in 1877, said nothing in reply.
At one o’clock, the bombardment began. Between 150 and 170 Confederate guns fired for nearly two hours, the largest bombardment of the war. Longstreet knew the distance. He knew what shrapnel did to formed ranks in the open. He knew from Malvern Hill that an artillery preparation could look successful from a mile away while being worth nothing at all. There was one thing about that battle he did not know. The bombardment was missing.
July 3 was hot and still. Smoke from 150 guns settled over the field and did not move. The Confederate gunners could not see where their shells landed, so they could not adjust. Shells flew long over the Union infantry and landed in the rear. They destroyed wagon parks. They hit reserve batteries. They hit Meade’s own headquarters hard enough that the Army of the Potomac’s commander had to pack up and move, while the men those shells were aimed at lay behind their wall waiting for the bombardment to end.
Henry Hunt had about 80 guns to reply with, and he was not going to waste them. He wanted every shell for the infantry he knew was coming. So he ordered his guns to stop firing. Not all at once. That was the trick. Hunt made his batteries fall silent one by one, unhurried, with pauses between them. The same rhythm a gun line makes when it is being dismantled piece by piece by a victorious enemy.
Through the smoke, Alexander watched the Union fire fade and die battery by battery, and read it the only way possible from his position. The impossible interpretation Longstreet had warned him he might have to make. The enemy’s ammunition was running out.
At 1:25, he sent a message to Pickett. The enemy’s fire had not slackened enough yet, but come on quickly or I cannot support the attack at all. Fifteen minutes later, a second message. The guns in front of him had fallen silent. Ammunition was nearly exhausted. Come on now.
Notice what that second message claims and what it does not claim. Eighteen guns had stopped firing, but Alexander had already told Longstreet he could only judge through counter-battery fire across smoke obscuring the entire field. “Withdrawal” was his best interpretation of the silence, not a confirmed count of anything. The sentence carrying the real meaning was the one that followed. Ammunition was nearly exhausted. That deadline was the true engine of the decision, not evidence that Union defenses had broken.
Pickett mounted his horse and rode toward the lines, holding Alexander’s note in his fist, to ask whether he should advance. What happened next has more than one account from people who were there, with no obvious reason to lie. Longstreet remembered, in his 1896 writing, that he tried to answer but failed. “The effort to speak the order failed, and I could only indicate it by an affirmative bow.” Alexander remembered, in his 1877 writing, that he heard no words and saw no bow at all, only a silence Pickett took as answer enough. The two accounts do not agree on the nature of the gesture. They agree on everything essential. The second-highest-ranking officer in the Army of Northern Virginia was asked a direct question in front of witnesses on the most important afternoon of the war, and no clear words came out of him.
Pickett mounted and rode down the lines. At three o’clock, about 12,500 men stepped out of the treeline, flags raised, bands playing, dressed in parade-ground formations across open farmland. Union artillery tore the ranks with shrapnel and airbursts, the ammunition Hunt had spent two hours refusing to use. The ranks closed the gaps and kept walking. Their objective, the copse of trees and the Emmitsburg Road, drew them into a crossfire. Pettigrew’s brigades collapsed on the left under enfilading fire. Pickett’s Virginians angled right toward the corner. Lewis Armistead led the final rush with his hat on his sword, the same Armistead who had carried his brigade’s signal up the slope at Malvern Hill and brought it back down.
For a few minutes, Confederate soldiers stood inside the Union lines, the farthest point the Confederacy ever reached, measured in yards and minutes. Then Hancock’s Second Corps closed around them from every side, 30 minutes of vicious fighting at zero range. Armistead fell mortally wounded, his hand on a Union gun. More than 6,000 Confederate soldiers were killed, wounded, or captured. Nearly 1,100 of them never rose from that field. Two-thirds of Pickett’s division did not return. Every one of his brigade commanders was killed or wounded. The division Longstreet had spent two days requesting was lost in 20 minutes, and Pickett never forgave anyone for it as long as he lived.
As the survivors straggled back, Lee rode out to meet them. “All this has been my fault,” he told them. “I have lost this battle.” Longstreet watched it happen from the hill exactly as he had predicted. Three separate times, to the only man who could have stopped it. There was no satisfaction in being right.
The retreat began the next day in cold rain and deep mud. Meade let Lee’s army leave rather than risk destroying it. The three days at Gettysburg cost both armies 51,000 combined casualties. On the same day the Confederates retreated south, Vicksburg fell in the west, splitting the Confederacy in two.
Lee, at least, was honest about his role. He took the blame himself in writing in his official report. He never claimed he had not been warned. He acknowledged the decision and the error were his own. Three weeks after the battle, in a private letter to his wife, he went further than any report required. He told her the army had labored hard, acted nobly, and accomplished all that could reasonably be expected of it. Then he added another line, quieter than anything in the official record. It was not to be expected that it would accomplish miracles or meet the expectations of the rash and unreasonable.
That is as close as Lee ever came, even in private, to admitting he asked his men for more than they could give.
It should have ended there. Instead, in the postwar years, Lost Cause writers, led by Jubal Early, rewrote what happened and blamed Longstreet. They claimed he was sluggish, that he deliberately delayed, that he disobeyed an order to attack at sunrise on July 2, an order for which no evidence exists that it was ever issued. The mechanism was elegant in its own way. You cannot blame Lee. Lee is the center of gravity of the Lost Cause, and blaming his military leadership risks blaming the cause itself. So you blame the man who disagreed with him instead.
It helped that man became a Republican after the war, supported Reconstruction, wrote his own account of Gettysburg rather than leaving Lee’s untouched, and publicly stated the South had been wrong about slavery, something no other senior Confederate officer was willing to say. The irony is precise. A man destroyed for something he did not do by people who knew exactly what he had done. Thinking clearly about this battle while it happened, and thinking clearly about slavery after the war ended. Honesty was uncomfortable in both directions.
There is a second thing the same myth needed to erase, and it belongs to this story too. The Lost Cause needed Lee’s army to be a specific thing, disciplined and honorable, beaten only by bad luck and Union numbers, not by anything the army itself did. Longstreet’s foresight threatened that story from one direction. A general who saw the disaster coming does not fit a narrative built on flawless leadership. What happened to free Black Pennsylvanians during the march north threatened it from another. An army that kidnapped people to return them to slavery does not fit a narrative built on Confederate honor. Neither truth was ever refuted. They were simply ignored. One man’s reputation was deliberately distorted to protect the myth. One atrocity was simply never included in the story. Different mechanisms, the same goal.
That generation did not stop at its battle or at that omission.
You may have heard the story that Longstreet’s family was so shattered by grief they could not arrange their children’s funeral, that a friend stepped in and handled the whole thing, burying the children himself. And that the friend was George Pickett. It is a beautiful story. It comes from a single source, a condolence letter written in 1904 on the occasion of Longstreet’s death by Pickett’s widow. A woman whom one of the most respected researchers in Confederate history concluded fabricated or embellished much of what she published under her husband’s name.
The truth is in the Hollywood Cemetery records, and no one put it in a book because there is nothing beautiful in it. Those three children remained in a vault borrowed from a stranger for eight years, alongside 29 other unclaimed bodies. They stayed there through the rest of the war. They stayed there when Richmond burned. They stayed there through Appomattox. In 1870, the cemetery finally emptied the vault, and that June, James Longstreet bought a burial plot for $28 and buried his children himself. Eight and a half years after the first of them died.
None of this settles every question this campaign still raises. Historians still argue about how much terrain slowed July 2, how slow Longstreet was to execute an order he hated, and how much autonomy he actually had on July 3 once Lee’s plan was set. Those are legitimate disagreements, and nothing here claims otherwise. But those arguments are different from disobedience. Disobedience is the charge that does not hold.
What actually happened? A general who objected clearly three times through proper channels, then executed the order he objected to, was never a scandal. It only became a scandal when postwar defenders of the Confederacy needed one, and had to assign it to someone other than Lee. Longstreet’s own defense never changed from his earliest writings on the subject to the end of his life. “I made my objections to Pickett’s charge, and they were overruled.” The casualty numbers, Lee’s own report, and the ground itself all eventually lined up to support his position. It took historians a century to get back to where he left it.
He lived until 1904, a diplomat and railroad commissioner, an old man who lived long enough to see everyone who stood with him on that hill pass away. He never got past those three days in Pennsylvania.
He knew. He knew from Fredericksburg what open ground did to men who crossed it. He knew from Malvern Hill what happened when you trusted an artillery bombardment to do an infantry’s work. He knew by the evening of July 2 that that specific position would not break, not from hope, but from his own casualty reports. He said all of it aloud three times to the only man on earth who could have acted on it. And when the moment came, 12,000 men crossed that field on a nod that no one fully agrees on the meaning of. Because the words, whatever they were supposed to be, never came.
Being right, in his view, had no power to prevent the thing he was right about. That is what is worth considering even now. More than a battle refought or a reputation restored. A considered, specific warning delivered three times through proper channels, which turned out to be the only account in this entire story that the evidence never had to catch up with.