Introduction

One of the challenges facing archaeologists of the twenty-first century is conducting archaeology in urban areas (e.g., Cantwell and Wall 2001; Kelly 1999; Lovis 2004; Rothschild and Wall 2014; Staski 2008; Yamin 2008). Not only does urban archaeology offer the technical challenges of excavating through deep fill or multiple deposits, but there is also the problem of sites that have either been destroyed by removal or covered by extensive development of the urban landscape. For archaeologists interested in spatial questions such as reconstructing prehistoric and historic use of landscapes, urban settings are sometimes a black hole compared to rural settings that facilitate relatively easy and widespread survey.

Archaeologists routinely turn to historical records like county histories, atlases, General Land Office survey notes, and antiquarian publications to find and document sites as a prelude to or in conjunction with archaeological research (Collins and Molyneaux 2003:33–34). In contrast to these “standard” sources employed by archaeologists, newspapers historically have been overlooked, or at least underutilized, as a documentary source of archaeological information. Ironically, one of the primary reasons for this is the local nature of newspapers. It might be assumed the local focus of many newspapers would make it likely that information on an archaeological discovery would be recorded in print, but any given state has had hundreds of newspapers, physically preventing archiving at most facilities including state historic preservation offices. This is especially true given that the value in storing newspapers would not be worthwhile in relation to the potential archaeological information contained therein. Rathje and Murphy (2001) hint at the scale of this storage problem when they note that a single year of the New York Times weighs about 520 pounds and takes up 1.5 cubic yards. The local nature of newspapers has also meant that physical access would be generally limited to towns where the newspaper was published with the result that examining these papers would be beyond the scope of many archaeological projects, particularly in counties with few libraries or small historical societies where newspapers might be kept and where hours or location might be inconvenient. Finally, as local products, newspapers tend to be irregularly indexed, if indexed at all, making searching for archaeological information not very cost effective for most projects. Even if indices exist, it would be rare for other than the most prominent sites (i.e., public historic sites) to be found in them since newspapers would be unlikely to neither name or use site numbers in their reporting.

The Digital Age has changed the potential for the utilization of newspapers as a reliable and effective source of archaeological information. Digitization of newspapers has almost entirely mitigated problems that hampered their use in archaeological research. Historic newspapers, in digital format, are now available globally via the internet. Furthermore, while perhaps not indexed, many digitized newspapers have been subjected to optical character recognition and thus can be searched for text, greatly easing the ability to find specific search terms. Because of this ease of access and general availability, digitized newspapers might now provide a valuable useful research tool for archaeologists who are looking to find new sites or to reconstruct the location of sites and settlement patterns. Since most urban areas had at least one daily newspaper, if not more, the possibility that newspapers contain information about sites long destroyed or buried has to be considered. This article proposes a general methodology for how to approach newspaper archives for the use of archaeologists and reports on the effectiveness of this approach utilizing Allen County, Indiana and the city of Fort Wayne as a case study (Fig. 1).

Fig. 1
figure 1

Diagram of Fort Wayne area showing concentrations of new sites: 1. Kekionga, historic native American village; 2. Wells’ preemption; 3. Original plat of Fort Wayne; 4. Swinney Park; 5. Near Richardville Reserve (Dots represent sites, lines represent streets with sites, and polygons represent parks or larger areas containing sites). Inset: State of Indiana showing the location of Allen County and its principal city, Fort Wayne

Full size image

Case Study: Allen County, Indiana

The idea to develop this methodology emerged from work on a previous project that relied on the use newspapers for background archaeological research. For an archaeological project at Hartman Reserve Nature Center in Cedar Falls, Iowa (Gaff and Caldwell 2012), it was discovered that John Hartman, for whom the property was named, was also an avocational archaeologist who collected in the area. Also revealed was that Hartman was the owner and editor of the local newspaper and that he had used his position and paper to increase awareness of the prehistory of the area. The most notable example of this was his lead article “Traces of Algonkian Culture” that was published in 1928. Contained in this article was not only a written description of finds from his decades of exploration but also a map of the Cedar Falls-Waterloo area with sites—mounds, excavated mounds, camps, and villages—denoted. This map was of such detail and quality that it was helpful for understanding how prehistoric inhabitants used the locality as it supported a more detailed and nuanced reconstruction of the prehistoric settlement system (Gaff and Caldwell 2012). Additional searches of the local newspaper produced several reports about finds from the area that enhanced the background historic research. After using newspaper research for subsequent projects and continuing to find value in it, the decision was made to develop a methodology for approaching newspaper research in a more systematic way and test it at the county level.

Allen County, Indiana, is the choice for this case study. A primary reason for choosing this county is that it has a large urban area, the City of Fort Wayne, with a long history of newspaper publication. This city, while large, is wholly contained within the county unlike some other larger cities that cross county boundaries which might complicate their use as case studies. Also anticipated was that the moderate size of the community, while allowing for a greater chance of finding appropriate newspaper articles, would not result in an unmanageable number. Based on the author’s prior participation in a variety of projects in the area, it was clear Fort Wayne represents a good case study of a growing urban environment whose landscape has been filled, sometimes to great depths. Such filling would result in the obliteration of sites through destruction or deep burial with the result that newspaper articles could potentially document sites that are no longer visible on the surface or that no longer exist. Finally, the county and city has a long documented history with fairly well-known prehistoric and historic Native American populations that have been studied by archaeologists, providing context for interpreting sites reported in newspapers.

In brief, Allen County is located in northeast Indiana and falls within three physiographic units, the Steuben Morainal Lake Area, Maumee Lacustrine Plain, and Tipton Till Plain (Schneider 1966). Most dominant of its geographic features are three rivers—the St. Joseph and St. Mary’s conjoin to form the Maumee. The Maumee flows into Lake Erie, and a short portage from southwest of Fort Wayne connects these rivers to the Wabash River. Thus, Fort Wayne, known colloquially as Three Rivers, is at the junction of the Mississippi River drainage and the Great Lakes, so this river system and the its resources are likely the reasons the area has rich evidence of human interaction extending far back to prehistory (Jeske 1990). Throughout the county is evidence for Archaic and Woodland sites, and recent research has shown that late prehistoric societies participated in broader regional cultural traditions (Carroll 2013).

Historically, European settlement of the area began with the construction of a French fort which was later replaced with a British fort. Today’s city, however, really begins with the military campaign of Anthony Wayne against the Northwest Territory Indian confederacy (Gaff 2004). The American fort that was built as a result of that conflict came to bear Wayne’s name as did the city that grew around it. After the Indian population was subdued and the War of 1812 brought to an end, a land office was established on May 8, 1822 with the first sale of land taking place on October 22, 1823 (Brice 1868). That same year, Allen County was organized. Perhaps the most important event in the early growth of the city was the arrival of the Wabash and Erie Canal. Building of that canal and its feeders led to a boom in both population and economic development that took Fort Wayne from frontier outpost to a full-fledged city before the Civil War (Poinsatte 1969). Following that war, the story of Fort Wayne is similar to that of other Midwestern cities with the introduction and growth of railroads, including interurban lines, fueling rapid expansion of light industry and manufacturing that characterized the economic profile of the city until after World War II (Ankenbruck 1975). Today, the city hosts a mix of service, retail, and manufacturing with a 2013 population of 256,496 and a county population of 363,014 in 2013. Since the growth of Fort Wayne from frontier settlement to modern city is similar to many other cities, it makes a good case study for using newspapers to find otherwise unreported archaeological sites, especially since it has nearly 200 years of development in its urban core.

Not surprisingly for a city with such a long history, a significant portion of the downtown area contains fill. Being at the confluence of three rivers meant there was originally much low, unusable land. Typical of many urban areas, early residents filled these areas with soil and rubbish in order to create more usable space above flood levels. The author participated in one project that involved testing in advance of the construction of Headwaters Park in downtown Fort Wayne (Jeske and Stillwell 1994). During this testing, fill was found to a great depth and while not recorded, an intact porcelain toilet was observed in a profile at a depth of over a meter. Another project at Guldlin Park also demonstrated the presence of fill (and the utility of newspaper articles), confirming news stories about infilling that particular location with coal (Gaff 2007, Journal Gazette, 26 February 1911, Fort Wayne News, 2 June 1914). As a case study for using newspapers to relocate these kinds of sites, it was expected Fort Wayne likely had numerous archaeological sites near the rivers that have subsequently been covered with such fill.

In addition to being a good example of a city with lots of fill, Fort Wayne has had a long history of newspaper publishing, thereby making it a good place to carry out a project involving them. The first paper, The Sentinel, started publication, with a Democratic leaning, in 1833. A Whig newspaper soon appeared in 1841. Other papers followed including several German newspapers (Ankenbruck 1975). Even with the newspaper industry rapidly transforming due to market pressure from internet news sources, Fort Wayne continues to have two main newspapers, the Journal-Gazette and the News-Sentinel. Such a long history of publication coupled with the fact that the Allen County Public Library has managed to collect most of the city’s papers (Beatty 2000), means that Fort Wayne is an ideal place to carry out this kind of case study.

Methodology

Searching local newspapers for archaeological sites has been greatly simplified in the twenty-first century by the digitization of newspaper collections and the availability of these via online databases. For this project, two databases were used—Newspaper Archives (www.newspaperarchive.com) and Fold3 (www.fold3.com). Both of these are available by subscription, but since many local and university libraries are subscribers—either directly or through other bundled subscriptions that include these databases—access to newspaper collections frequently can be secured at no cost and without the need to travel to examine physical archives. The Fold3 database emphasizes military records and contains copies of newspapers as ancillary to that goal, whereas Newspaper Archive focuses exclusively on newspapers, claiming to be the largest such collection in the world and adding a newspaper page per second to its database. In this particular case, Fold3 provided a research advantage in that it had access to the newspaper collections at the Allen County Public Library and had taken the extra step to make the scans searchable. One limitation for databases such as these is they generally only make available newspapers that are in the public domain. For this study then, newspapers only as recent as 1923 were used, although a search via another website, www.godfrey.org, yielded two articles from 1933 (Fort Wayne News Sentinel, 16 May 1933, Fort Wayne Journal Gazette, 17 May 1933).

Fort Wayne is a large urban center with a long history and a population that increased greatly after the completion of the Wabash and Erie Canal. Given such a large population by the middle of the nineteenth century, it is not surprising that Fort Wayne and the surrounding communities supported several newspapers over the years, and this was reflected in the database search. Listed in Table 1 are newspapers that ended up employed in this study. Other local newspapers were investigated. But since they did not contain relevant information, they were not included.

Table 1 Titles and dates of publication for newspapers used in this case study. Publication information from the Library of Congress’s websites, Chronicling America: historic American newspapers available at http://chroniclingamerica.loc.gov/
Full size table

Being aware that newspaper reporting of accidental discoveries, especially prior to World War II, would not include the exacting scientific terminology of today’s modern archaeology (e.g., terms like Mississippian, flotation, or survey, etc.) the challenge was to utilize a set of terms that would have been in common usage one hundred years ago and general enough to locate most of the relevant articles, while specific enough to minimize extraneous articles and increase the overall efficiency of finding stories concerning possible archaeological sites. The first attempt to search for articles did not meet with much success because the terms employed in the search were too specific. For example, “Miami” for one of the groups that occupied the area in the historic era and “Kekionga” the main historic Native American settlement in the area produced very few articles describing archaeological finds.

Much more productive was to use the search term “Indian,” especially when coupled with additional search terms. Using this somewhat generic word did result in a larger number of articles, but owing to its generality and the nature of how the newspapers are digitized and organized, it also increased the number of false positives. One common example was to search for “Indian” and “burial” where the result from such a search would include a newspaper page with Indian motorcycles (a popular motorcycle company in the first half of the twentieth century) and an obituary or funeral notice. The reason for this—and a key point constraining these searches—is that most newspapers are digitized by page, not by article. Thus, multiple search terms can pull from different articles to yield a result such as that described in the Indian motorcycle and obituary example. Because of this issue, for every relevant newspaper article located, there were roughly five “non-articles” produced as search results.

Terms that would be common in pre-World War II newspaper reporting, utilized in conjunction with “Indian,” included “skeleton,” “remains,” “skull,” “bones,” “grave,” and “burial” to find burials. Also used were the terms “burying ground,” “cemetery,” and “graveyard.” Finds of artifacts were more difficult to search for but “trinkets” used with “Indian” frequently yielded results as did “artifacts,” “relic,” “curio,” and “arrowhead.” Excavation terms used with “Indian” were “dug up” and “unearthed.” “Cellar” and “sewer” were two other words used to find accidental discoveries. Finally, the words “chief” and “warriors” were used with “Indian” as this was a common designator for describing Native American bodies found in burial sites. While not used for this project, in retrospect, “princess” would have likely been a good word to use in searches because it turns out this was a common designator for the remains of female Indians based on newspaper articles encountered.

After carrying out these searches in the databases, the articles were printed and roughly sorted into those that might provide useful archaeological information and those that would not. Articles with little utility for archaeological studies tended to be write-ups of local history that mentioned Native Americans like Miami Chief Little Turtle as opposed to documenting local archaeology (e.g., Fort Wayne News, 24 June 1911, Fort Wayne Sentinel, 28 February 1914, Fort Wayne Journal Gazette, 12 July 1914). Stories of this type were disregarded. This preliminary sort left well over 100 articles that might be useful for determining the location of archaeological sites, so the creation of a database served to record the articles as well as to abstract information from them to facilitate sorting and searching.

A database created in Microsoft Access was designed to highlight the archaeological character of the articles. Roughly, the original database contained four basic types of information: the newspaper source, geographic location, persons involved, and nature of the find. For the source of information, the database included fields for Newspaper Date, Newspaper Name, and Page Number. Since reporting on geographic information was highly variable, there were several fields for location data including: Street Number, Street Name, Property Name (used for things like parks or local landmarks), Block or Intersection, as well as a field for Other Location Information. This particular variable was used for additional geographic details like references to older historic events (like battles or camps), specific buildings on properties like farms or estates, or the names of businesses. For persons involved there were two fields for Land Owner and Discoverer. When it came to the nature of the find, two Yes/No checkbox fields were created—one for the presence or absence of burials and one for the presence or absence of artifacts. Associated with the Burial column were fields for whether it was a single inhumation or a multiple burial. For the Artifacts field, additional supporting fields were created for lithics, ceramics, metal, and other artifacts. Finally, a field was made for Other Information to contain data that would not fit in otherwise and for notes related to decisions about populating the database.

After the development of this initial database, the geographic information contained in the database was used to plot the sites on street maps of Fort Wayne and Allen County (for sites falling outside the city limits). After this initial plotting on paper maps, production of final maps involved using Adobe Illustrator to add sites to imported digital USGS topographic maps. If sites were plotted, they were added to a Yes/No field that was added called Plotted. These plotted sites were then compared to the State of Indiana’s SHAARD (State Historic Architectural and Archaeological Research Database) and the SHAARD GIS in the summer of 2014, both maintained by the Department of Natural Resource’s Division of Historic Preservation and Archaeology. SHAARD contains recorded archaeological and historic sites in Indiana; sites reported in the newspaper articles that did not appear in SHAARD were noted in an additional Yes/No column labeled New Site.

The final step of the research involved a survey of sites documented in newspaper sources. Potential site locations that could be plotted were visited by driving to the locations and visually inspecting them. Investigation of locations in rural areas outside the city was by simple drive-by while sites concentrated within the city, particularly downtown, were checked as pedestrian survey. All these locations were categorized into several general types. Urban sites were those locations that were completely developed and thus not likely amenable to further field investigation short of major mechanical removal owing to the presence of structures like parking lots, office buildings, or strip malls. Sites labeled as “residential” were in neighborhoods with houses and yards with a modest potential for easy access and investigation through excavation. Sites with the best possibility for testing were those classified as “farm” (or “rural” for wooded property) or “park.”

Results

Ultimately, the number of newspaper articles discovered that contained enough geographical and archaeological data about finds to possibly warrant the designation of archaeological site were entered into the database and totaled 121. A little less than half of these (n = 54) ended up considered as archaeological sites. Since the articles were read, evaluated, and entered into the database regardless of date of publication (i.e., due to the nature of the search, articles were not recovered in or entered into the database in chronological order of appearance in the newspaper), there was a problem with duplication. Seventeen of the articles entered into the database were duplicate news stories. These duplicates reflected the same news story published in two different newspapers, usually published within a day or two of each other. However, worth noting is that duplicates of this sort still retained some utility because a different write-up of the same event might include different details or provide information missing from other accounts.

Another, and perhaps more interesting, kind of duplication was different finds from the same specific area or archaeological sites. One of the outcomes of this project was the identification of a few archaeologically rich areas in Fort Wayne. During the course of this research, there were many instances encountered where the newspapers reported on discoveries from a particular area and subsequently described additional finds in the same place. For sites with multiple finds, the location was noted on a map, and then other articles about that location would be noted as having been recorded earlier. In the database, there were another 17 newspaper articles recorded as redundant in this way.

In addition to duplicates, the other major classes of articles not used were those that had issues with their geographic information. Originally, they were included in the database because they contained some geographic data upon initial reading, but when actually plotting sites, these proved to be more difficult to place for one reason or another. One issue for these articles were locations where the geographic information was too general for documenting archaeological sites. Typically, these concerned longer streets in the city where the article did not provide an owner, street number, or other unique identifier. For example, several articles described finds on “West Washington” which stretches one and one-half miles from downtown Fort Wayne west to the St. Mary’s River. Classifying the entirety of West Washington Boulevard as an archaeological site did not make sense, even though several finds came from there. Sixteen of the newspaper articles had information that was too general for use according to these criteria. It should be noted that sites reported as coming from streets that were a few blocks or less in length were treated as archaeological sites given the much smaller area involved.

Another major class of sites (n = 8) in the database that were not recorded were those that included geographic information, but that would require meticulous research of property records beyond the scope of this project. One example is the “Anderson property” described as being “a short distance south of Lawton Park” (Fort Wayne Sentinel, 3 October 1903). In a case like this, known elements include the property owner, date, and some geographic detail. Obviously, terminology like “south of Lawton Park” is vague as it does not have any measure of distance. Resources utilized to track down properties for this project include county histories, local atlases, Sanborn fire maps, and plat books. These did not show the Anderson property. However, the kind of information found in the Anderson property article would permit one to likely eventually track down a property with additional research in tax, census, or deed records. Similarly, farms named after their owners that fell outside of the references consulted for this project present a similar kind of problem.

Comparison of the project data with SHAARD showed that a few sites had either already been recorded from historic records or investigated in the field by archaeologists. Since these had been recorded in SHAARD, they were discarded from consideration as newly found sites. An example of this is 12AL12. The newspaper article for this site was a relatively more comprehensive article by a well-known antiquarian, R.S. Robertson, who wrote: “descending the St. Joseph on the east, to the farm of Peter Nutestine, we find a circular fort or earthwork situated in the bend of the river” (Robertson 1895:27). Robertson’s report of sites is common knowledge, and the expectation was that all of the sites in it would have been investigated and recorded. This was not always the case, but in this particular situation, the earthwork—the Adams Enclosure—had been investigated in 2008 (McCullough 2009). Similarly, 12AL26 was “rediscovered” through newspapers and removed from the list of possibly new sites.

In a few other instances, a newspaper article site was recorded in SHAARD, but for an entirely different reason. One case of this is the burial of Little Turtle, war chief of the Miami. Little Turtle’s remains were uncovered in Dr. George W. Gillie’s lot on Lawton Place in 1912. Currently, the burial site is maintained as a small memorial park in a city lot sandwiched between two houses. Prior to and after the discovery of Little Turtle’s remains, the newspapers reported that several different skeletons were found on the property at different times (Fort Wayne Journal Gazette, 19 May 1911, Fort Wayne Journal Gazette, 16 April 1916, Fort Wayne Sentinel, 4 April 1921, Fort Wayne Journal Gazette, 13 April 1921). So, while the site is already recorded as a burial site for Little Turtle, the archaeological site is likely much larger than originally recorded based on the number of finds documented in the newspapers. Another case like this is the Swinney property. Swinney was an early and prominent settler in the Fort Wayne area, and he maintained a large estate west of downtown. As the newspaper accounts indicate (Fort Wayne Weekly Gazette, 11 July 1895, Fort Wayne News, 24 July 1895, Fort Wayne Daily Gazette, 25 July 1895, Fort Wayne News, 4 October 1902, Fort Wayne Journal Gazette, 15 November 1914), the broader area around his home is dense with archaeological sites and the park that bears his name, and the nearby land surely contains several sites. One newspaper (Fort Wayne Sentinel, 10 September 1900) described a burial with grave goods consistent with an American Indian buried in the historic era. These remains were found near the Swinney residence, which is still standing today. Because the house still stands and is operated as a historical museum, it appears as a historic property in SHAARD. Sites like these, where the newspaper article hinted at an additional archaeological components different than what was recorded in SHAARD, were not counted as new sites. Archaeological sites that were associated with more contemporary buildings without an obvious historical or archaeological connection to any potential remains (e.g., those under the recent expansion of the Allen County Public Library or the former city-county building, now known as the Rousseau Center) were considered to be new archaeological sites. Finally, one interesting article described a set of bones, silver cross, and bells on a piece of cloth recovered from a load of fill dug out of a gravel pit deposited at a another location (Fort Wayne Journal Gazette, 16 September 1906). That material was not recorded as a site, although the gravel pit was since a find there appeared in a different article.

Following the removal of articles with duplicative or insufficient data from the database, this methodology produced a total of 57 new site locations. Of these, 46 were burials, including both single and multiple burials. The locations included two of the city’s first cemeteries. The remaining 11 sites were those containing only artifacts or those in which the description find was described as Indian relics where it was not known whether the term “relics” simply referred to artifacts or also included human remains. The large number of burials relative to artifact sites is, in part, the result of a methodology that leaned more towards using search terms involving burials. However, the greater reason for this bias is the nature of newspaper reporting. Finding skeletal remains, especially those tied to the early history of the city or the area’s aboriginal inhabitants, clearly made for more sensational news stories than more modest finds of just artifacts. While there is no way to know for sure, most of the burials seem to date to the historic era. This is based on the observation that when artifacts were noted with the bodies, they are items like crosses, beads, and other items suggestive of trade goods. Similarly, clothing items like epaulets indicate an age contemporary to the American invasion and establishment of the fort.

Bias also appears in the types and kinds of artifacts reported. Scientific surveys of the St. Joseph and the St. Mary’s rivers show that prehistoric ceramics exist at sites in Allen County (Jeske 1992, 1996). From the newspaper stories, however, there are almost no specific mentions of prehistoric ceramics. Again, this could be the newspapers’ tendency to use sensational stories. It might also be a reflection of the community’s standard at the time in terms of what was interesting or valuable since tomahawks and pipes receive frequent mention as do silver crosses and beads. In the articles examined, there are only 11 instances of reporting stone materials, of which only 4 were arrowheads, the remainder being stone pipes and axes. This again seems to reinforce the idea that reporting of spectacular finds appear more often than those of mundane artifacts.

The drive-by survey to assess the current archaeological potential of the 57 newly recorded sites confirms that most of these are deeply buried or completely lost and that the newspaper articles might be the only source of data about these specific locations. Twenty-six of the locations were classified as urban. These sites were mostly concentrated in downtown Fort Wayne and include places that are under existing parking lots or buildings (i.e., large office buildings, factories, or shops) or under the massive dikes that form part of the city’s flood control system. For the most part, these might be considered lost since excavation at these locations is virtually impossible. Sites classified as residential were those predominated by single family dwellings with yards where houses are organized into neighborhoods. Twenty sites fell within this classification. Like sites classified as urban, excavation might be challenging giving the presence of structures and buried utilities. On the other hand, since most of these neighborhood sites represent original construction (i.e., did not undergo multiple episodes of construction and destruction like the urban sites), it is conceivable that excavation in yards and empty lots might be productive. Finally, nine sites were located in either rural and farm settings or in city parks. Presumably, the potential for these sites to yield information from archaeological investigation is high. Regardless, these results confirm that most of these sites are inaccessible and that the newspaper articles might be the best and certainly most readily available source of information about them.

History and geography strongly structure the distribution of sites. By history, it is meant that a bulk of the newly defined sites are located in or very near downtown Fort Wayne. Given the time period of the newspapers used, this makes sense owing to the fact that most of the civic growth and accompanying excavation was happening in that area and that suburban migration had not yet occurred. So, with most of the development being in downtown, it is not a surprise that this is where most of the discoveries that took place. Also, the density of occupation in the inner city would more likely result in a downtown find being reported as opposed to one from the rural countryside for a variety of reasons including that newspaper reporters would have had more ready access to city news. History, of course, also plays a role in that most of the key historical events took place near downtown Fort Wayne.

Geography also seems to structure the distribution of sites. The expectation is that the three rivers would host prehistoric and historic Native American populations and this is borne out by the fact that a majority of these sites are near these primary waterways. The ones found outside city limits are located on creeks and along the St. Joseph River. None of the sites in the surrounding county were along the Maumee or St. Mary’s River. Within the city proper, with the exception of a few outliers, there are several discrete clusters of sites, some of which reflect the known historic use of the area by Native Americans (Gaff 2006). The first of these are the sites recovered northeast of the confluence of the rivers. The density of sites here is not surprising since this is the location (Poinsatte 1976) of the historic Indian village of Kekionga as well as the British fort (Fig. 1, group 1). The cluster of sites between Spy Run Creek and the St. Joseph River falls within the area known as Wells’ Preemption, where Indian Agent and Miami adoptee, William Wells owned property and likely let some Miami reside (Fig. 1, group 2). The cluster of sites just south of the confluence of the rivers is likely related to the fact that this specific area was the first plat of Fort Wayne and so is where the earliest historic development would have taken place (Fig. 1, group 3). Sites in and around Swinney Park do not correlate with any historically documented Indian camp or village (Fig. 1, group 4). Finally, a small cluster of sites along the St. Mary’s in the southwest part of the city corresponds to property owned by Jean Richardville, civil chief of the Miami who owned a large reservation in this area (Jeske 1995) and is reputed to have hosted an Indian settlement in addition to his immediate family (Fig. 1, group 5).

Overall, the results of the newspaper research help fill in gaps in the archaeological record. With the history of development in Fort Wayne, it is obvious most sites in downtown are, for most practical purposes, inaccessible either by being covered with fill and buildings or by having been hauled away during the evolution of the urban space. So, research in newspaper archives helps to fill in the archaeological history of the area. This is especially important since most of these sites were not in SHAARD at the time of this study. The primary reason being that the bulk of the newspaper sources came from between the 1870s and 1920s, prior to the formalization of archaeological site recording at the state level. Also, as indicated above, newspaper stories about accidental finds have not traditionally been used as a source of archaeological data. From these two factors—the date when finds happened and reporting as news, not science—it is easy to see why so many of these sites would not be incorporated into state site files. The Fort Wayne case study documented here demonstrates the utility of recovering information about long-forgotten sites in urban contexts from newspaper research. This sort of research also appears to be valuable in identifying archaeologically rich areas for further testing because even though locations in the downtown area are in urban context, excavation in nearby contexts like vacant lots might be productive. Similarly, in more rural areas, and perhaps, some neighborhoods, the methodology has also resulted in identifying a small number of relocated sites suitable for further exploration through excavation.

Potential Uses for the Methodology

Adopting a methodology of trying to systematically search newspapers for articles about accidental finds and others sorts of archaeological sites would be beneficial for several different kinds of archaeological projects. Primary amongst these would be projects that involve recreating past settlement systems and those investigating site distributions and use of landscape. Graduate students and professional researchers engaged in such research can, with a modest amount of effort, potentially add dozens of sites to their state inventories. Not only would this lead to more robust models by increasing the overall number of sites, as indicated above, it might also lead to sites or archaeologically rich areas that would be worthy of testing.

Along this same line, those involved in this kind of research might consider using this methodology as the basis for undergraduate research projects or independent studies. The basic steps involved in this approach—finding articles, plotting them on maps, comparison with existing site inventories, and field investigations—do not require a great deal of effort. Conceivably, such work could be easily accomplished over the course of a semester and summarized in a paper or report. The advantage to archaeologists working with undergraduates is that using this for a research project, because of its obvious and intuitive nature, would not require lengthy or difficult training as can sometimes be the case with teaching students to do more specialized kinds of analysis.

Finally, there is obvious value in doing this kind of work for state historic preservation offices and other government agencies tasked with inventorying sites and maintaining databases for sites. Addition of sites found by searching newspapers would contribute to the overall knowledge about site distributions and, thereby, enhance the management of cultural resources. The reason for this is that not only does the methodology increase the overall number of sites but it also potentially adds components to sites previously recorded as single component. Another benefit of adopting this methodology, not only for state agencies but researchers as well, is that it could be a source of funding. As an example, state historic preservation offices could seek grants to fund a search of newspapers for all the counties in their respective states. Such funding could pay for everything from newspaper database subscriptions to paying staff to survey newly found sites. Funding could also cover the data entry of the hundreds of sites that would likely be discovered. Finally, worth remembering is that most of the newspaper databases only make available papers that are in the public domain. As that date continues to roll, every five to ten years, researchers could repeat the process, adding even more sites as newer articles become available.

Regardless, the reporting of accidental finds and other archaeological information documented in historic newspapers certainly has the potential to add data to a variety of archaeological projects. This information is vitally important for urban areas, like Fort Wayne, where many of the sites have been filled or destroyed, and it is likely the newspaper articles are the only, or certainly most accessible, evidence of these sites. The case study illustrated here demonstrates that just using these newspaper sites alone allows patterns of settlement and landscape usage to emerge. Combining this new source of archaeological information with existing site data makes the newspaper information even more useful.