On a blustery afternoon in March 1864, a horse-drawn carriage pulled up in front of the Washington studio of famed photographer Mathew Brady. From it emerged Ulysses S.Grant, recently appointed commander of all Union armies, and Secretary of War Edwin M. Stanton, who understood very well the value of giving the country a good look at its new top general.
Brady ushered the pair into his “operating room,”where the president and other dignitaries frequently came to have their likenesses made. Glass skylights brilliantly illuminated the three cameras that dominated the room. A trio of assistants scurried quietly about their preparations. As Stanton looked on, Brady positioned Grant in front of the cameras. The general stood with his left hand tugging the lapel of a knee-length military coat. Two assistants got behind cameras; Brady himself manned the third. He reminded Grant to remain absolutely motionless for at least 12 seconds, required by the wet-plate exposure process.
Just as the session was about to begin, the room suddenly darkened: A cloud had passed over the sun. To compensate, Brady sent an assistant to the roof to pull a canvas mat from one of the skylights. The assistant took a misstep. The skylight shattered. And in one horrible instant a deluge of glass, heavy and razor-sharp, fell all around the Union hero. A dagger-like shard struck Grant in the neck, cutting the carotid artery. All color drained from his face. He collapsed, still feebly clutching his lapel. Brady and Stanton instantly rushed to his side, but there was not a chance of saving him. Within moments the victor of Fort Donelson, Shiloh, Vicksburg and Chattanooga lay motionless in a pool of his own blood.
The above vignette is fiction, but according to one account, it missed becoming fact by as little as one-sixteenth of an inch.“It was a miracle that some of the pieces didn’t strike him,” Brady insisted. “And if it had, it would have been the end of Grant; for that glass was two inches thick.” Brady told the story to play up Grant’s imperturbability— “the most remarkable display of nerve I ever witnessed.” But the incident also shows how easily a key historical actor can be removed from the scene.
Such “what ifs” are a common feature of Civil War literature. MacKinlay Kantor, in his classic If the South Had Won the Civil War (1960), also killed off Grant—in a freak equestrian accident at a crucial juncture of the Vicksburg campaign, thereby placing the Army of the Tennessee under its senior corps commander, the incompetent Maj. Gen. John McClernand. McClernand led that army to disaster, while in Pennsylvania General Robert E. Lee’s Army of Northern Virginia destroyed the Union Army of the Potomac and captured Washington, D.C. (a hapless Abraham Lincoln included).
Unsurprisingly, Gettysburg is by far the most popular target of “what if” history. Peter G. Tsouras’ Gettysburg: An Alternate History (1996) imagines how that epic struggle would have unfolded if Maj. Gen. J.E.B. Stuart had reached the battlefield a day earlier, if the Confederates had captured Cemetery Hill on the evening of July 1, if Lt. Gen. James Longstreet had conducted a much wider envelopment of the Union left flank on July 2, if Pickett’s charge had been better conducted and if the Federals had launched an immediate, vigorous counterattack. In Gettysburg:A Novel of the Civil War (2003), Newt Gingrich and William Fortschen also rewrite the battle to kick off the first of three novels about the Civil War’s final years. And as long ago as the 1930s, a famous double negative “what if”essay appeared, looking back from a fictive world in which the Confederacy was an established nation and Europe was prosperous and at peace. Titled “If Lee Had Not Won the Battle of Gettysburg,” it imagined a dire future in which carpetbaggers despoiled a defeated South, a resurgent United States picked a fight with Spain at the turn of the 20th century and a chain of events ensued that led to a calamitous world war barely a decade later. Its author: British statesman Winston Churchill.
Although fun to read and often well-informed and accurate on specific details of weaponry, tactics, personalities and so on, these alternate histories are all but useless in terms of enhancing our grasp of the Civil War. The principal reason is their blithe piling of one speculation atop another. Even if each speculation had a 90 percent chance of being correct—so that the probability of the first divergence from historical reality was nine out of 10, as was the next turn of events in the new narrative and the next— then after just a dozen such turns, each with an individual likelihood of 90 percent, the probability of the resultant “alternate world” would be only 0.282. In a major battle, 12 important turns of events could take place within a day, perhaps even a few hours.
But a growing number of scholars are coming to embrace a more rigorous form of “what if” thinking. Known as counterfactual history, these thought experiments are valued mainly for two reasons: They force the historian to think more carefully about causal claims, and they prompt questions the historian might otherwise not think to ask.
Good counterfactuals are not the same animal as the casual “what if” questions discussed over beer and pretzels or written principally as entertainment. They instead abide by rules of logic that have been hammered out over time by scholars in several disciplines: philosophy, psychology, economics, history and political science. An excellent recent example is Roger L. Ransom’s The Confederate States of America: What Might Have Been, published in 2005.To be sure, many historians—and some laypeople—still exhibit a knee-jerk rejection of counterfactual analysis. But any history that asserts a causal connection between events has within it an embedded counterfactual. Thus if one were to argue that Lincoln won reelection because of Sherman’s capture of Atlanta in September 1864, the argument contains the implicit counterfactual that the failure to capture Atlanta would have cost Lincoln the presidency.
The case of the “fatal” photography session at Brady’s studio is intended to test a part of the traditional Civil War narrative so familiar that most people, scholars and buffs alike, take it for granted. It is the trope that frames the Union’s major command problem in terms of Lincoln’s search for a general capable of winning the war. This is the theme of numerous classic works, notably T. Harry Williams’ Lincoln and His Generals (1952) and Kenneth P. Williams’ multivolume Lincoln Finds a General (1949, later expanded).
By the end of 1863, of course, Lincoln had found his general. Thus the standard accounts of the Civil War’s final year invariably begin with Grant’s arrival in Washington on March 8, 1864; and then his planning the great spring campaign of 1864—a plan that included visits to Maj. Gen. William T. Sherman in Nashville and Cincinnati, Maj. Gen. George G. Meade in Culpeper, Va., and Maj. Gen. Benjamin F. Butler at Norfolk and Fort Monroe.
The centerpiece of the plan was a series of two major and two subsidiary offensives all aimed toward the heartland of the Confederacy and timed to begin simultaneously, to prevent the Confederate high command from shifting its forces from one threatened point to another. And once the Federal offensive began in early May 1864, Grant supervised the various Union armies to an unprecedented degree, especially the hard-luck Army of the Potomac.
Implicit in this narrative is a claim made explicitly in such works as Robin H. Neilland’s Grant: The Man Who Won the Civil War (2004): Grant was indispensable to Northern victory, and without him the war for the Union would have been lost or at least greatly protracted. Plunging a counterfactual shard into Grant’s neck forces the historian to rethink this claim— in effect, to put pressure on the causal variable of Grant’s generalship and oblige it to work for a living.
Removing Grant from the historical equation raises several questions a historian might otherwise never think to ask. The first and most obvious is, who would have replaced him? In all likelihood, the job of general-in-chief would have reverted to Grant’s predecessor, the cerebral Maj. Gen. Henry W. Halleck, nicknamed “Old Brains.” He at least was a known quantity. Sherman, Grant’s handpicked successor in the Military Division of the Mississippi and the candidate most likely to occur to us today, would almost certainly not have received a presidential summons to come to Washington and assume the mantle of his fallen friend and partner in command. Sherman had only led an army in a single battle—Chattanooga in November 1863—and although the battle as a whole ended in Union triumph, Sherman’s share of it had not gone well.Major General George H.Thomas, whose Army of the Cumberland played the decisive role at Chattanooga, hailed from a seceded state, Virginia; he was neither well known to Lincoln nor likely to have received Halleck’s endorsement. In any event he could not have been promoted without snubbing Sherman. Meade, for his part, was simply out of the question. Despite his great victory at Gettysburg the previous July, his performance afterward had disappointed Lincoln. Even worse, in March 1864 he was still enduring an inquisition from Congress’ Joint Committee on the Conduct of the War, thanks to persistent claims that Gettysburg had been won in spite of his generalship, not because of it.
Would Halleck have adopted Grant’s strategy—still in embryo at the time of his notional demise in Brady’s studio—of a series of simultaneous Union offensives? Almost certainly not. During his 19 months as general-in-chief, he had never proposed such a thing and indeed had a principled belief that only commanders in the field should be the final arbiters of when and how to conduct their operations.
This raises the question of the importance of the simultaneous offensives. Just how vital was this component of Grant’s strategy, and how serious would its absence have been? Richard McMurry, himself a practitioner of counterfactual history, has argued that it was not vital at all and proved a flat failure when actually implemented: “Never before or since the spring and summer of 1864 did the Confederates shift troops about so freely, in such large numbers, and with such success as they did during those months.”He goes on to give no fewer than five examples to prove his point. As for Grant’s notion of subsidiary offensives to support the two main offensives by Sherman and Meade, every one of them collapsed within two weeks of the opening of the spring campaign. At least two of them—Maj. Gen. Franz Sigel’s fiasco in the Shenandoah Valley and Ben Butler’s Bermuda Hundred campaign—arguably did more harm than good.
What then of Grant’s generalship once the campaign began? Grant gave Sherman a great deal of autonomy in the Western theater, so it is likely that the Atlanta campaign would have followed its historical path whether or not Grant lived long enough to see it unfold.But in the East, Grant kept Meade on so tight a leash that reporters quickly dubbed the Army of the Potomac “Grant’s army.” How well would Meade have performed without Grant’s supervision?
In essence, the answer to that question depends on how one responds to three component questions: First, was the Grant-Meade relationship an effective partnership? (If not, Grant’s near-continual presence with the Army of the Potomac might in fact have had an adverse impact.) Second, could Meade have borne up psychologically under the pounding Lee would have given him in the Wilderness? (Under Grant, the Army of the Potomac absorbed as many casualties as it did at Chancellorsville in May 1863.Major General Joseph Hooker, who led the army in that engagement, abandoned his offensive and withdrew. Grant famously persevered and ordered the army south toward Spotsylvania Court House. Which course would Meade have adopted?)
Finally, assuming that Meade persisted as Grant did, could he have duplicated Grant’s feat of achieving a 10-month stalemate in the Richmond-Petersburg area? (It is hard to imagine his doing any worse.Between the Wilderness and Cold Harbor,Grant attempted an almost endless stream of attacks on Lee’s army but could never get the Army of the Potomac to move as fast as he desired. The chronic results were attacks made too hastily or too late, resulting in 55,000 casualties and nearly wrecking the army as an effective offensive instrument.)
The case of the fatal photography session asks us to reconsider not only Grant’s generalship but also the depth of the Union’s reservoir of military talent at the senior level. If it was sufficiently deep—if generals such as Sherman, Meade, Maj. Gen. George H. Thomas and Maj. Gen. Philip H. Sheridan could have won the war—then Grant was in effect dispensable. If it was not sufficiently deep, then it is worth asking why it was not.
The level of Union military talent is an example of a second-order counterfactual. That is, having constructed a minimal rewrite of history to produce a departure from what actually occurred, a responsible historian must consider other factors that would have tended either to intensify the departure or, more usually, to produce a return to the original historical path. First order counterfactuals, like Grant’s untimely death, often focus on important individuals, for obvious reasons. “Many explanations in History and the social sciences,” writes Geoffrey Hawthorn,“turn not on causal connections between states of affairs that are beyond human control, but on the relevant agents’ own practical reasonings.” Eliminating a powerful “relevant agent” like Grant is an easy way to alter history. In contrast, second-order counterfactuals usually emphasize structural considerations that are hard to alter in any plausible way. Often these turn out to be “reversionary” counterfactuals that effectively nullify the impact of the first-order counterfactual.
For example, even without Grant, the North would still have held a commanding materiel edge over the Confederacy by virtue of its larger population, manufacturing capability, financial resources and so on. These factors would have enabled the Union to keep larger forces in the field—while the Confederacy would still have faced serious fissures within its society. Would these and similar factors have produced a Northern victory despite Grant’s absence?
How far should a historian attempt to press a counterfactual thought experiment? Some believe it is possible to peer decades or even centuries into the future. In The Confederate States of America, for example, Roger Ransom feels comfortable extrapolating the future of an independent Confederacy as far as 1914. “History is rarely, if ever, like coin tossing,”explain the editors of Unmaking the West: “What-If?” Scenarios That Rewrite World History (2006); “rather, it is a path-dependent system with positive feedback. It resembles a vortex in which what has already happened quickly accentuates the possibility of certain events and reduces that of others, making escape from the new path difficult.”
Others are more conservative. They believe that the more one presses a counterfactual, the more artificial and therefore unhelpful the exercise becomes.By this reasoning, after attempting to glimpse the near-term consequences of the Union’s having to wage the Civil War without Grant’s aid from March 1864 onward, the historian must return to Brady’s studio on what was in reality a not so fateful afternoon.The accident occurs, the glass shards fall all around Grant, but they leave him unscathed.The photography session proceeds, producing some of the best images of Grant we have. Stanton sternly warns Brady to say nothing of the accident lest it spark rumors of an assassination attempt. He then escorts Grant from the studio to a train that will carry him to his headquarters in the town of Culpeper, just across the Rapidan River from Lee’s waiting army.
Mark Grimsley, a pioneer in the use of counterfactual history, is a professor at The Ohio State University and author of the acclaimed The Hard Hand of War: Union Military Policy Toward Southern Civilians, 1861-1865.
Originally published in the May 2007 issue of America’s Civil War. To subscribe, click here.