How to Discover Soldier Photos from a Specific Civil War Battle

By Kurt Luther 

How can we find photos of Civil War soldiers connected to a particular battle? Earlier this summer, I was confronted with this question when Emma Nostheide, a historian at Manassas National Battlefield Park, contacted me. She was developing a museum exhibit, “Faces of the Fallen,” that would showcase portraits of Union and Confederate soldiers who were killed or mortally wounded at the Battles of First or Second Manassas. She aimed to include over a hundred such images and sought the help of the photo sleuthing community to expand her collection. I told her we would be happy to help.

Most of the time, photo sleuths focus on researching a particular soldier or photo. Less often, they investigate a group or collection. Seeking to identify photos of participants in an entire Civil War battle, especially a major one, is relatively rare, and the process for doing so wasn’t obvious to me. Yet I could imagine other parks or museum staff might have similar goals. I decided that, while I worked on this Manassas-centric project, I would also try to document my process such that other researchers could adapt it relatively easily for other battles.

Of course, I wasn’t starting from scratch. Nostheide had already researched the battles and generated an initial list of 76 names and units. About two-thirds of this list participated in Second Manassas or both battles; the remainder participated in only First Manassas. I decided to scope my efforts to First Manassas and try to add as many names to Nostheide’s list as possible through my own research. Then I would share our combined list with the broader community and invite others to pick up the torch.

The Battle of First Manassas, also known by Northerners as First Bull Run, was the smaller of the two battles that took place around the same area of Northern Virginia. But, as the first major battle of the Civil War, it brought together two substantial forces on July 21, 1861. According to the American Battlefield Trust, Brig. Gen. Irvin McDowell’s Army of Northeastern Virginia, with about 28,000 Federals engaged in the battle versus about 32,000 Confederates under the combined forces of Brig. Gen. Pierre G.T. Beauregard’s Army of the Potomac and Brig. Gen. Joseph E. Johnston’s Army of the Shenandoah.

When the battle ended in a Confederate victory, there were over 4,800 casualties, including over 800 dead, according to the same source. The Official Records of the Union and Confederate Armies (OR), published in the late 19th century, breaks down the numbers as 481 Union dead (19 officers, 462 enlisted) and 387 Confederate dead (25 officers, 362 enlisted). It is impossible to know how many of them had their photos taken during their lifetimes, much less how many of those photos survived today and could be properly identified. Nevertheless, Nostheide and I agreed it was worth a try.

My research followed a straightforward three-step process: identifying the soldiers killed at First Manassas, searching for their photos, and comparing my findings to the existing NPS list. Since I wanted my process to be generalizable, I would initially prefer comprehensive sources for photos and military records over battle-specific ones.

Identifying soldiers killed or mortally wounded in a battle

Surprisingly, there is no single, centralized, official source listing individual names of soldiers killed in Civil War battles. For example, the National Park Service’s Civil War Soldiers and Sailors System (CWSS), available for free online, provides searchable information about soldier names, ranks, and regiments, but not where (or whether) they were killed in action. Likewise, Fold3, a paid subscription service from Ancestry.com, lists the soldiers and regiments involved in each battle through its “War Stories” feature but doesn’t allow filtering by those who died. Many states’ Adjutants General kept detailed records of their units’ casualties, and some of these are digitized and searchable online, but those that include specific names often do not break them down by battle, or vice-versa.

The best source I could find that includes specific names of soldiers killed for all major battles was The American Civil War Database (HDS), a paid subscription site. HDS aggregates data from many primary sources, including the aforementioned Adjutant General reports, but its true power comes from linking these records together in a relational database and making them searchable. For a given regiment, one can view a list of all battles in which it participated, and then drill down to see the names of individual soldiers killed, wounded, or captured there. Unfortunately, the reverse is not yet possible – one cannot select a battle and see a list of names for all of its casualties.

The workaround, an admittedly much more manual process, is to search each regiment involved in the battle on HDS and then view its list of casualties for that battle. This requires finding an order of battle, which lists all the regiments that participated in a battle, who commanded each, and how they were organized. There are several good sources for orders of battle. The OR contains many, and searchable copies of every volume are available online for free via HathiTrust Digital Library. In addition to orders of battle and the aforementioned casualty list, broken down by regiment, the OR also contains after-action reports submitted by each unit commander that may include a few names of soldiers killed or wounded, especially officers. However, they often do not, and such lists are usually incomplete, so additional sources are needed.

If the battle has an affiliated unit of the National Park Service, its website also often contains an order of battle under the “Learn” > “History & Culture” tabs. Finally, Wikipedia contains over 200 articles with Civil War orders of battle, many of which cite the aforementioned references, but double-check them because anyone can edit a Wikipedia article and there may be errors.

Of course, for a momentous historical event like First Manassas, there are also battle-specific resources containing orders of battle. My favorite battlefield guide series, This Hallowed Ground by University of Nebraska Press, includes orders of battle in the appendices of every book, including Manassas by Ethan S. Rafuse, which I referenced frequently for this column. Two of the best digital history websites for First Manassas, including detailed orders of battle, are Harry Smeltzer’s Bull Runnings and Jonathan Soffe’s FirstBullRun.co.uk. Across these diverse sources, I found the orders of battle contained minor variations (e.g., assigning an artillery battery to one brigade versus another) but were largely consistent.

Using the order of battle, we can now search each regiment on HDS to find the names of individual soldiers who were casualties of a given battle. However, finding the right unit in HDS can be trickier than expected, especially for early-war battles like First Manassas with many single-digit regiment numbers. For example, I often found three or four versions of the “1st Infantry” for the various states that participated in First Manassas. The record-holder was “1st Arkansas Infantry” which had no fewer than seven variants: one battalion, two late-war “consolidated” units, one “state troops,” one “militia,” one “volunteers,” plus the one I was actually looking for. I also found that artillery and cavalry units are especially difficult to track down. In particular, HDS struggles to differentiate the different batteries of the U.S. Regular Army artillery units, as well as Confederate cavalry regiments with uniquely named companies.

I found two useful techniques for overcoming these issues. One approach was to open up each of the HDS candidate results in separate browser tabs,” click the “Regiment Casualty Analysis” tab for each of them, and see if there is an entry dated around July 21, 1861, with the place name as either “Manassas” or “Bull Run.” If the place name appears with casualties greater than zero, you know you’ve found the right one. But sometimes, especially if the unit suffered no casualties, the battle name doesn’t show up at all.

A second approach was to start by looking up the unit commander’s name in HDS. Each soldier’s profile typically lists all the units he served in, so from there I could click on the matching regiment and pull up its casualty lists. This approach, while slower than the first one, was often more successful, but even this wasn’t foolproof. Sometimes HDS linked a soldier to the wrong unit. It also requires identifying the commander’s name, which is not quickly found in the OR; some reports were filed by unit commanders, others by superior officers or subordinates. Furthermore, sometimes the name is wrong or misspelled; my searches for “J Seymour” and “W Featherson” proved fruitless until I realized they were typos on the NPS website. Thus, triangulating across multiple orders of battle can actually save time.

As I worked through all units in the First Manassas orders of battle–56 Union and 68 Confederate–I soon discovered that HDS’ records, while an excellent start, were far from complete. I found names for 259 of the 481 Union soldiers reported killed in the OR, along with 212 of 387 Confederates. In total, I was missing 397 names of fallen soldiers, so my odds of finding their photos were already cut in half.

Locating photos of the fallen soldiers

There are many places where one can find photos of Civil War soldiers (see MI Summer 2015, “The Photo Sleuth’s Digital Toolkit”). Since I wanted to use a generalizable process, and I had over 450 names to look up, I limited my photo search to three platforms: HDS, Civil War Photo Sleuth (CWPS), and Find a Grave. Given that the intended purpose of these photos was a museum exhibit, I preferred photos that had verified, trustworthy identifications and those that were held in public collections, which would be easier to get permission to use. Again, for generalizability, I did not consult battle-specific sources (e.g., books about First Manassas) or regimental histories, although these are valuable references.

HDS, according to its home page, contains over 26,000 Civil War soldier photos. Most of these come from the U.S. Army Heritage and Education Center’s MOLLUS-Massachusetts public collection, as well as subscriber submissions. The easiest way to find relevant ones for a battle search is to simply click the links for each soldier name listed under the relevant regiment’s casualty analysis for that battle. If there is a photo, it will appear on the left side of the biography page. One can also view all photos for a particular regiment, but without being able to filter just those killed at First Manassas, this is less efficient.

CWPS contains over 60,000 Civil War soldier photos from a mix of public and private collections. As CWPS focuses on photos, it has limited biographical data on specific soldiers. It doesn’t know if the soldier was killed during the war, much less which battle, so there is no way to filter by these criteria. The best approach depends on how many fallen soldiers there are for a given regiment. If that number is small, i.e., five or less, it’s faster to simply search each soldier’s first and last name on CWPS, one at a time, and see if there is a result.

If there are more than five names to search, it’s more efficient to search CWPS for all photos for a given unit. This can be tedious, but I found a setup that is relatively streamlined and reduces errors. I started by searching the regiment name and number on CWPS. (If there was no regiment number, e.g., Hampton’s Legion, I just searched by state and narrowed things down manually.) Next, I scrolled through all the photo results until I got to the very bottom of the page. Then, I used my browser’s “Find in page” feature to search the entire page for each surname on the HDS list–just the first few letters is usually enough, but I had to watch out for multiple matches. My computer screen was wide enough that I could speed things up by putting the HDS and CWPS browser windows side-by-side. HDS defaults to 20 results per page, so I made sure to advance to the next page if that regiment had more than 20 fallen soldiers at First Manassas. This approach yielded 8 new photos out of 471 names searched.

I wanted to try one more photo source: Find a Grave. Owned by Ancestry.com since 2013, Find a Grave is an online database with information on over 230 million memorials, including photos of many burial sites and decedents. Find a Grave also allows users to create “virtual cemeteries” that group together disparate graves in meaningful ways, such as a particular regiment or casualties of a battle. Find a Grave pages are user-generated content with minimal quality controls, so as with Wikipedia, researchers must bring a cautious eye and verify claims. But sometimes, these pages can yield diamonds in the rough: photos that cannot be found anywhere else.

My strategy on Find a Grave was to first search for individuals who had served in the military and died around the time frame of First Manassas, and then focus on the subset that also had photos of the decedent. On the home page, I entered “1861” for the death year and “US” for the cemetery location. (I didn’t pick “Virginia” because some soldiers may be reinterred.) I also ticked the “Veteran” checkbox. These search criteria yielded over 5,600 results. Next, I changed the sorting drop-down menu from “Relevance” to “Death Date (Older)” so that the results would be ordered by the decedent’s death date, starting with Jan. 1, 1861. I started scrolling and eventually – there were over 200 pages of search results – made it to late July.

Scrolling more slowly now, I inspected the thumbnail for each search result, looking for portraits rather than headstones. This approach is not foolproof, as the thumbnail could be any of the photos attached to a memorial page, but usually the portrait is featured as the thumbnail if there is one.

Verifying and comparing results

I found 26 candidates with portrait thumbnails between the dates of the First Manassas Campaign, between July 16-22. Of these, 11 were already on the NPS list, six were unrelated to First Manassas, and one I had already found during my HDS search.

The remaining eight photos represented new soldier photos not seen previously. Knowing Find a Grave can be an unreliable source, I reviewed these results carefully. One was clearly wrong: a photo of Joseph R. Anderson was misidentified as Ebenezer K. Fleming, colonel of the 1st Virginia Infantry, who was killed at First Manassas. Another photo’s identification was described as just a guess. The final six photo IDs were all possibly or definitely correct. That is, they represented new photos of soldiers who had fallen at First Manassas.

Overall, my search across the three platforms uncovered 14 new photos: eight from HDS and CWPS, and another six from Find a Grave. Emma Nostheide, the Manassas historian who researched the original NPS list, was grateful to receive them. “The contribution of the images [you] sent are an integral part of a larger communal effort that has aided the park in its effort to interpret the human cost of these battles,” Nostheide wrote to me. “Overall, photo projects like these are a great way to help visitors connect with the park by showing them the human element of the story.” The Faces of the Fallen exhibit is planned for a Memorial Day 2025 opening at the battlefield’s visitor center.

Opportunities for further research

Even when we add the roughly 50 First Manassas photos that the park staff had already found, the total is less than 10 percent of the 868 soldiers known to have fallen there. Yet, I’m optimistic there are many more photos waiting to be found. The process I followed provides many opportunities for more fine-grained follow-up research.

One promising path would be to find battle-specific books and regimental histories to identify the names and units of the nearly 400 fallen soldiers my first pass missed. Another would be to search public collections, such as MOLLUS-Mass., and the web more broadly. For example, the OR reports that the 4th South Carolina Infantry lost 11 soldiers killed in action, yet HDS shows zero results. Where else might we find their names? On the Bull Runnings website, a letter written by Corp. Warren D. Wilkes of that regiment, published in the July 29, 1861, edition of the Wilmington (N.C.) Daily Journal, gives the names of several others, including Orderly Sergeant J. W. Morrice (“Morris” in HDS), Pvt. John Simpson, and his brother, Capt. Samuel Marion Wilkes.

Furthermore, we have not even touched Second Manassas, a much larger battle with over 1,700 Union and 1,000 Confederate soldiers killed. The park staff is eager to receive any photos connected to these fallen soldiers, as well.

Thus, our final technique is crowdsourcing, reaching beyond the limits of individual research. We invite you, fellow photo sleuths, to submit any First or Second Manassas photos of fallen soldiers you may own or be aware of that we haven’t found yet. An updated list will be maintained for the next few months. You can also email me at kurt.luther@photosleuth.org (use the subject “Manassas Faces of the Fallen”) and I will make sure it gets forwarded to the park staff, with full credit to you.

I hope the research process described here is useful, not just in surfacing more faces of the fallen at the Battles of First and Second Manassas, but also for future photo sleuths hoping to discover the names and stories of soldiers who participated in other battles.

Kurt Luther is an associate professor of computer science and, by courtesy, history at Virginia Tech and an adjunct professor at Virginia Military Institute. He is the creator of Civil War Photo Sleuth, a free website that combines face recognition technology and community to identify Civil War portraits. He is a MI Senior Editor.

Call to action

Do you have, or are aware of, photos of fallen soldiers from First or Second Manassas not on our list? If you do, email me at kurt.luther@photosleuth.org (use the subject “Manassas Faces of the Fallen”) and I will make sure it gets forwarded to the park staff, with full credit to you.

SPREAD THE WORD: We encourage you to share this story on social media and elsewhere to educate and raise awareness. If you wish to use any image on this page for another purpose, please request permission.

LEARN MORE about Military Images, America’s only magazine dedicated to showcasing, interpreting and preserving Civil War portrait photography.

VISIT OUR STORE to subscribe, renew a subscription, and more.