New digital tools developed for history help to provide new ways to analyze historical data. These new ways to analyze data not only help to answer questions but also undoubtedly raise new questions. Like the many other graduate students who have posted on this blog, I applied one of these new digital history tools to the information wanted ads of the Christian Recorder. More specifically I used new mapping tools to map the locations associated with these information wanted ads and found it raised more questions than it answered. But I am getting ahead of myself. Let me explain a bit about the project we all undertook and the Christian Recorder.
The Christian Recorder was one of the most popular African-American newspapers during the Civil War, though publication started nine years before the war in 1852. It was run by the African Methodist Episcopal church and was not only used for religious and church news but also for regular news and advertisements. The newspaper still runs to this day. During the later stages of the Civil War, many ads under the heading of information wanted started appearing in the newspaper. These ads were posted by African-Americans looking for long lost family members and friends. A majority of these people were separated by the effects of slavery, whether it was being sold away from one another, people running away from slavery, or one of the hundreds of other ways people could be separated by slavery or in an active war zone. Even after the war ended, these ads continued to be posted in the Christian Recorder. These ads provide a unique opportunity for historians to get more information on a group of people who are very under represented in primary sources from this time period.
As part of a project for a digital history class at Villanova University, my classmates, my professor and myself made a huge excel project incorporating information from these ads posted between late 1864 and summer of 1869. In this excel project we tried to include information about the people posting the ads, termed the Advertiser for this project, (name, gender, relationship to the person they were looking for, reason and length of separation, and mailing address for the ad) and the people being looked for in the ad, termed the Advertisee for this project (name, gender, place of origin, last known address, and relationship to the Advertiser). Besides this information pulled from the ads, some rows had additional notes about points of interest from that particular ad. With all of this information, the amount of research, and digital projects that could spring from it are too many to count. I chose to do a project using a digital map making program. I chose to do this because this whole semester I have enjoyed learning about map making software, whether it was map warping or some other map making software.
This leads me to try to figure out the best way to apply mapping software to this chart and which software to use. As to the second part of my quandary I chose to use Google Fusion Tables. Google describes their new software as, “an experimental data visualization web application to gather, visualize, and share data tables.” Not only can you upload your own excel files or use your own files on Google Drive but if you just want to experiment with the program you can use other people’s data to play around with. As described above, besides making this data easier to share, it allows you to map the data in Google Maps. As with most things with Google, even though this one is still classified as experimental, it is free, easy to use, and impeccably well run.
Once I figured out which program I was going to use, I determined the best way to use this program was to map the location of the subscribers to the Christian Recorder. I chose this as my original map because of a fundamental question that developed in my mind. The question was, how wide spread was the circulation of the Christian Recorder? Depending on how wide spread it was, it would certainly influence how effective the information wanted ads would be at finding the people in the ads. Since our project covers about five years and I only had a short period of time to assemble this information, I knew I would have to choose only one year. I chose 1866 since it was the year that had the most ads posted in it, according to our research. Each issue of the Christian Recorder had a section that would normally list people who had either just renewed or just subscribed to the paper, called acknowledgments. I went through each issue in 1866 listing the locations of these subscribers. I did not record the names of the subscribers since I was studying the locations of the subscriptions and not the people who were subscribing, which would be an interesting study in and of itself. After going through the paper I was able to assemble the following Excel spread: Christian Recorder Subscribers 1866.
Once I had assembled the spreadsheet I uploaded it to Google Fusion Tables to map the different locations. Since the places I was trying to map were relatively straight forward, just city, state, and country, I thought it would be an easy process. Little did I know it would be like trying to pull teeth from an angry tiger. When I tried to map the locations originally it put them all over the map. So to fix this I thought I would find the latitude and longitude for each location. This was no small undertaking since I had 274 rows of data. There were some repeats in the subscriptions but not that many. When I tried to map these locations, it totally failed again (user error I learned later). After some choice words to the computer, I looked in the help section and figured out what I was doing wrong. I had to combine the city, state, and country into one cell. After figuring this out, I was then able to actually map the locations of the subscribers for 1866, which you can see pictured below. When you pull up the window, click where it says map to view the map. Also since there are a large number of data points, it does take a a bit of time for all of the data points to load in. (Repeat for the next maps)
After making this map I came to a conclusion that a majority of the subscribers for 1866 to the Christian Recorder were in the North and Upper South. This did not surprise me to any real extent since literacy rates among African-Americans was probably higher in the North as well as the infrastructure would be more in tact in the North since it was not torn apart by the war like the former Confederacy. What I think is also interesting are the subscribers located far out west in Colorado and California. It would be interesting to know how those people got out there, but that is for another project. Yet, this map didn’t really answer how effective these ads would be at finding people that were in the ads. To do this I decided I should map the last known locations of the Advertisees. Making this map went much faster since I knew how to make the Google Fusion Tables recognize locations in a table. After making my custom table with just the last known locations of the Advertisees and uploading it to Google Fusion Tables, I was able to produce the following map.
As would be expected, the last known locations of most of the Advertisees was in states where slavery was either legal (the border states) or practiced (Confederate States of America) until the end of the war. When compared to the subscribers map, you can see that there are large gaps in the former Confederacy. There are some subscribers in the South but when you get to the Deep South the subscribers are few and far between. Some states like Alabama, Florida, and Texas have no subscribers in it. This would make the likely hood of the ads finding anyone in those states very low. Of course these are just the subscribers for 1866, there could be subscribers in those states before or after 1866. It also may be that the Christian Recorder didn’t publish all of their subscribers in the paper. This means there could be subscribers in those blank areas that are not listed in the paper. Also some regiments of the United States Colored Troop had copies of the Christian Recorder, and many of them were on occupation duty (even though it was not called occupation duty) in the South throughout much of Reconstruction. To try and map the movements of all the U.S.C.T. regiments from 1864-1869 would be a very hard task and beyond my scope. It is safe to say they some of the regiments would have copies of the newspaper as they traveled through parts of the Deep South where there were few to no subscribers. This would increase the likely hood of the ads actually finding the intended people.
After making the map of the last known location of the Advertisees, I came to a realization. Besides the map of the subscribers from 1866, I should map the locations of the Advertisers. I think it is safe to assume that the people posting these ads would have access to a copy of the paper, whether it was their own copy or a copy they borrowed from someone. It was common that churches would have copies of the paper for people to read. This would allow me to map the entire time period we researched where copies of the Christian Recorder would have been, according to our research. I made the Advertisers location map using the same method as the Advertisee map. I was then able to create a map of all the locations of the Advertisers.
The locations of the advertisers matches closely to the map of the subscribers but there are some differences. There are a few people down in the Deep South but more than the subscribers map and this map shows someone out in Idaho. Now these differences can be explained by using some of my theories I suggested earlier. The advertisers represent people from 1864-1869, which means their subscriber base changed over time. It also shows, when examined with the data table, that not all subscribers were published in the paper. To get a better overall comparision between the three types, advertisers, advertisees, and subscribers, I combined the three maps.
When all three maps are examined, you can clearly see that there is a lack of Christian Recorders in the Deep South. As discussed above, just because the map doesn’t show that there are copies down there doesn’t mean that they aren’t there. But even if there are more copies in the Deep South, we can assume they are few and far between since barely any show up on this map and if there was a prevalence of copies in the Deep South there would be more of them on this map. This raises the question as to why there are so few subscriptions in the Deep South? True it is a Philadelphia based paper but it does have an audience all the way out in California. Another question that comes to mind is why are there so many of the papers in the South located near the coast or a river? Are there large African-American communities there? Is it easier to transport papers there? Is there a larger presence of the Union Army there? This map really does raise more questions than it answers.
That is the curse and blessing of using digital history tools to analyze data. It is a blessing because it can help to answer questions about data. By mapping all of these different points you can see where subscribers, advertisers, and advertisees were located. It helps to show the coverage of the Christian Recorder and the information wanted ads. It shows that these ads probably were not very effective at finding people because of the coverage or lack there of, of the Christian Recorder when compared to the last known locations of the Advertisees. Then again it is a curse because it raises even more questions such as those mentioned above. Yet, this is what historians strive for, not only to answer questions but to raise new ones. This is how we as historians can push forward historical knowledge and learning.