New Feature for the Civil War Soldiers & Sailors Database
One of the most consistently useful online research tools available for Civil War folks is the National Park Service’s Civil War Soldiers & Sailors (CWSS) database. It has a number of features, but the most important is the database of soldier’s names and units, which can be searched a number of ways. If you’re looking for a specific Civil War soldier, it really is the place to start before going on to more in-depth primary sources, like the national Archives’ compiled service records (CSRs) at subscription sites like Fold3. One nice aspect of the NPS database is that it seems to be drawn from the CSRs at the National Archives (Mike, is this correct?), which means that if you find an entry in the database, there should be ay least something at NARA. As I say, it really is the place to start.
The CWSS recently had a big makeover, which improved its look tremendously. As for search functionality, I dunno. The original system had very much a Web 1.0 sort of interface, and it was clunky. The new system is clunky in a different way, with extensive pull-down menus that have to be opened and closed individually. This helps standardize queries much more precisely, and allows users to search on multiple parameters at once — say, both cavalry and dismounted cavalry — and that’s good. But it’s also a bit more complicated, maybe more than it needs to be — in designating the geographic origins to search, is it really necessary to use a two-tier arrangement that first requires one to select the letter, T, and then Texas from a second menu?
Nevertheless, that’s a minor quibble. One tool that seems to be new, that I’d not seen before, is the option of downloading an Excel file containing all the men in the database for a specific regiment or other unit. This is tremendously valuable , as it puts a more-or-less complete — or as complete as the records are — listing of an entire regiment’s complement at your fingertips with one click. The fields (columns) include name, initial rank, final rank, and company within the regiment. This is an example from the 1st Texas Heavy Artillery Regiment, that was stationed here during the war:
(This example also highlights something to be aware of in both the CWSS database and the CSRs from the National Archives; there are entries for each spelling of soldiers’ names in the cards at NARA. Privates Joseph Maschack, Joseph Maschick, and Joseph Maschuck are all pretty clearly the same individual.)
When searching for a regiment or similar-sized unit, the user will get a brief synopsis of that unit’s service, with links to the CWSS database entries for major actions in which they fought, like this one for that same regiment:
Overview: 1st Heavy Artillery Regiment was organized at Galveston, Texas, during the winter of 1861-1862 using the 3rd Texas Artillery Battalion as its nucleus. The unit served in the Trans-Mississippi Department primarily at Galveston and along the upper Texas coast. Company F was stationed at Sabine Pass during September, 1863, and was prominent in the surrender of two Federal gunboats, the Sachem and Clifton. In April, 1864, it was stationed at Galveston Island with 23 officers and 462 men, and in April, 1865, there were 430 present for duty. The regiment was included in the surrender on June 2. Its commanders were Colonel Joseph J. Cook, Lieutenant Colonel John H. Manly, and Major Edward Von Harten.
Then when you click the link “View Regiment’s Soldiers,” you get this:
Click on “Download Spreadsheet” and you’re in business.
What makes this so useful is that, once you have the Excel file, that can serve as the framework for a larger database, with other fields and information to be added from other sources — a scaffold of sorts, for larger research projects. The data can be sorted by rank or company, allowing one to quickly compile a list of (say) the regiment’s officers, or all the members of a given company.
This is a very useful feature of the CWSS that, I think, has not been available previously. It opens up a lot of new possibilities, and puts a powerful new tool in the hands of professional and avocational historians alike.
Good hunting, y’all!
Dead Confederates, A Civil War Era Blog 
Andy: You are correct that the database uses the Compiled Military Service Records at the National Archives. There are, of course, a number of deficiencies in it, some of which you mention in your post. Nevertheless, I have consulted it with some frequency since I retired and no longer have convenient access to the CSRs, and have found it quite helpful. We owe a considerable debt to the many gallant volunteers who worked on this project, and to those who designed and oversee it, including those from the National Park Service, NARA, the LDS, and other organizations. I anticipate that it will be added to over time.
Mike, thanks very much for that. I have been getting regular notices that Fold3, which makes CSRs available by subscription, is continuing to add Union records, at the moment from USCT regiments. These records are digital scans of the originals (as opposed to scans of older microfilm), so they’re in color. Odd how that small difference — which has little or no bearing on the informational content — seems to make them more compelling.
i’m not that familiar with these databases, so it’s interesting to learn about them. as an experiment, i looked for my ancestor Alexander Spotswood Thomson, for whom i have copies of records located by my grandfather Thomson many years ago. the simple search failed to turn up any Thomsons in the 2nd Virginia Cavalry, so i tested the regimental access — not grabbing the spreadsheet, but instead using the web interface. it’s pretty bloody awful, but after a bit of fooling around, i found A.S. Thompson: http://www.nps.gov/civilwar/search-soldiers-detail.htm?soldier_id=044842d9-dc7a-df11-bf36-b8ac6f5d926a
i don’t know where the incorrectly spelled last name was introduced, but the records for which i have copies get it right. i wonder how common these sorts of errors are?
Richard, those sorts of errors are very common, but they’re present across the board in genealogical research, from all sorts of sources. “A bit of fooling around,” as you say, is often required.
As mentioned, the database draws from the Compiled Service Records at the National Archives. These, in turn, were assembled by the War Department in the early 1900s, where vast numbers of clerks went through regimental rolls and other documents, month by month, year be year, and filled out a card for each man, for each document, transcribing key information. These cards were then assembled into folders, Compiled Service Records, that could be used as a quick refer4ence for each man’s service, at least as far as records survived. It was a tremendous undertaking.
In any event, any errors in name spelling in the original documents was carried over into the individual cards for the CSRs — note that when the cards were being compiled, the clerks had no means of cross-referencing them against other documents. In the case of the ACTUAL CSRs, these mistakes were often corrected by inserting a note directing the reader to another set, but that’s not reflected in the CWSS database.
You can download a copy of Pvt. Thomson’s CSR here. It’s short, only five pages, sporadically covering the last year of the war. It looks like he was admitted to military hospitals around Richmond in April and May 1865, no illness or injury specified. There is also a two-item CSR for one Pvt. A. G. Thompson, also of Co. G, 2nd Virginia Cavalry, noting hospitalization at this same time, that I believe refers to the same man, the middle initial S being confused for G.
if i’d really been thinking clearly, i would have looked at my grandfather’s collected documents before posting that. indeed, the name is spelled Thompson on the muster rolls, and the copies i have are of the same documents as appear in the CSR. so much for trusting to memory…
the broader question to me is why is the search so unusable? my consulting work since February of this year has involved learning a great deal about search engines and how they work, and the NPS database search engine is simply not making use of a lot of things that are standard in modern search technology. if you have a catalog of common spelling errors, you can easily account for that in a well constructed search index. wildcarding should be readily available (so i can search for Thom* easily.) i tried such a query with the NPS database and it doesn’t appear to be there. the NPS site forces only fairly structured searches; there isn’t really a good reason why an unstructured search (type in words and click search) can’t also be supplied side-by-side with the structured search.
Richard, I agree that the interface and searching is clunky. All I can suggest (and this is just a guess on my part) is that this online function is done in-house, with very little (or no) set-aside, dedicated resources, so they’re lagging well behind what one has come to expect in commercially-driven tools.
the thing is, i’m using open source tools (Apache Solr) and they’re way out in front of what i’m seeing at this NPS engine.
Let’s hope that someday we’ll have a database of white naval personnel. I don’t know about full-fledged CSRs, but enlistment logs do exist; Michael Bennet used them for his outstanding book “Union Jacks” on Federal sailor life.
“Union Jacks” is a great book, and I especially liked his coverage of African American sailors in the USN, a subject that often gets glossed over with generalities. Bennett makes clear that their experiences varied tremendously with the theater of the conflict they were in, especially on the rivers.
Yes, there’s a lot yet to be done in indexing and digitizing.