In physical space, where is the Mississippi River? This may seem like a relatively straightforward question, but from the perspective of a historian, I think a different question—What is the Mississippi River?—needs to be answered first. On a map, the location of the Mississippi seems relatively easy to identify as the blue line that roughly bisects the United States. Take into consideration the wider watershed and natural changes to the riverbed and water levels, and, on the ground, the river is not so easy to identify. Treating the river as a historical object adds more layers of meaning, specifically the layers of human experience. The historical Mississippi is the river recorded in the memories, photographs, newspaper articles, and newsreels of the people who have experienced it in the past. Tracking down the locations of these parts of the river—the conceptual river—was the goal of my research assistantship at the Institute for Advanced Study at the University of Minnesota during fall 2016.
My aim was to locate documents related to health in the Mississippi watershed, so I started my search with a map of the watershed. As I determined which search terms would be most useful for locating documents kept in archives across the United States, I decided to pair a subject term, like “malaria,” with a tributary term, like “Ohio River.” This helped me to search for documents related to events that happened in a broader swath of the watershed, but it also was imperfect. Rather than getting blanket coverage of the entire watershed, I still captured only documents that were already associated with the river, so if someone had malaria somewhere in the middle of Ohio, even if it was in the Mississippi River watershed, I probably would not have found records of it with my search. In this case, the “watershed” was a swath a few miles wide on either side of major tributaries.
This issue with capture was less of a concern with my most fruitful subject search, “floods.” More than one-third of the 1,500 document citations I collected are associated with flooding of one or more tributaries in the Mississippi watershed or the Mississippi River itself. Many rivers have flooded repeatedly in the same place: the Ohio River, for example, had major floods in 1884, 1913, 1936, and 1937. If you have any experience with the river, you are likely not surprised by this, but the repeated flood events add another relevant layer to my initial question: Where is the river? And where did the historical residents of Pittsburgh think the river was? The river is poorly represented by a line on a map, in part because it changes and moves. Sometimes, when the water is high, the Ohio River and the city of Pittsburgh are one and the same. Geographically speaking, where I find the river today may be very different from where I will find the river tomorrow, depending on whether the snow starts melting.
What stays the same is the place where the river was, say, in the spring of 1937. This definition of place, along with a record of impact, are the kinds of things the archival record preserves. According to one stylized map of the 1937 disaster, the river was much, much wider than a line, engulfing cities including Pittsburgh, Cincinnati, Louisville, St. Louis, and Little Rock. The map, intended to show the extent of response to the disaster, includes a visual representation of where aid boats, clinics, and supplies were deployed, along with the numbers of each. Along the Ohio and Mississippi Rivers, 12,721 square miles and 1.5 million people were affected. More subtly, the map also suggests why some places remained unaffected, by including the levees that edged the river south of Cairo, Illinois and, evidently, surrounded Vicksburg and Natchez, Mississippi.
Both immediate and long-term responses to floods are recorded in archival documents in the form of government reports, maps, oral histories, manuscript letters, and film reels. Documents related to the Mississippi watershed are distributed at archives concentrated in the watershed. The Minnesota Historical Society (MNHS) houses one particularly strong collection, in part because it includes both the state archives and those of the historical society. It is also well organized and well connected to national databases, and it contains thorough item descriptions and, frequently, extensive finding aids detailing collection contents. Of the 1,500 relevant documents I recorded, the 100 from the MNHS collection represent the largest holding of a single archive. The collections of the Linda Hall Library in Kansas City were also relevant to flooding on the Mississippi River, though less related to health. As a repository for documents related to engineering, their collections of documents related to the U.S. Army Corps of Engineers and flood control are extensive. In contrast to the topically organized collection of the Linda Hall Library, the archives at the University of Wisconsin-La Crosse has a river-related collection of a particular type of materials: oral histories. These oral histories were primarily collected between 1999 and 2007, but focus on life along the Mississippi River in the mid-twentieth century and cross topics from childhood memories of fishing and boating on the river to opinions about flood control. Together, the collections at the MNHS, the Linda Hall Library, and the University of Wisconsin-La Crosse represent a cross-section of the types of archives with relevant documents—government, private, and university archives.
The narratives of health in the Mississippi watershed are not confined to the Midwest and neither are the archival materials related to them. From the Huntington Library in Los Angeles County to the New York Botanical Garden in the Bronx, river-related documents appear in collections related to medicine and public health. The UCLA Film Archives offers another collection organized by material type, containing many newsreels that included flooding highlights from the 1920s through the 1960s. These and other archives demonstrate the interconnected nature of a history of the Mississippi River and show that archival information flows in different directions and amounts than the river itself. They also indicate that a raft trip down the Mississippi River would not get a researcher anywhere near many of the documents related to the history of the watershed, not only because many state and university archives are located some distance from the river, but also because the conceptual river is part of a much larger history of the United States dispersed in archival form across the country.
 Even this is not nearly so straightforward, especially near the headwaters. For one thing, the river is too narrow to show up on maps. In addition, geographic surveys are hard to complete and can be biased. See: http://www.startribune.com/minnesota-history-controversy-at-the-mississippi-s-headwaters/294522401/
 Of course, there were floods in many other years; these are just the ones best captured by the archival documents I recorded.
 All of the archival documents referenced in this piece are digitized and available online. http://library2.usi.edu:8080/cdm/ref/collection/1937_Flood/id/2190.
Juetten, Reba. 2017. “The Flow of Health, Water, and Information in the Mississippi Watershed.” Open Rivers: Rethinking Water, Place & Community, no. 7. http://editions.lib.umn.edu/openrivers/article/the-flow-of-health-water-and-information-in-the-mississippi-watershed/.
Download PDF of The Flow of Health, Water, and Information in the Mississippi Watershed by Reba Juetten.