Skousen and Aiuvalasit state that we do not consider the alternative hypothesis that the post-Mississippian peak in fecal stanols dated to AD 1650 could have resulted from the erosion of archaeological sediments containing fecal stanols that were initially deposited long before they entered Horseshoe Lake. Data are available to test this hypothesis. We used mineral content data and the age model from core HORM12 (Munoz Reference Munoz2015) to calculate Horseshoe Lake mineral accumulation rates, which measure catchment influx (White et al. Reference White, Stevens, Lorenzi, Munoz, Schroeder, Cao and Bogdanovich2019:5463). The mineral accumulation rate is near a minimum at the time of the AD 1650 peak, indicating that erosion from older archaeological deposits is an unlikely explanation for the increase in fecal stanols.

Skousen and Aiuvalasit cite Tse and colleagues (Reference Tse, Codling, Jones, Thoms, Liber, Giesy, Wheater and Doig2014) to suggest that “not enough is known about how stanols accumulate in lake basins” for us to interpret the fecal stanol record at Horseshoe Lake. The study by Tse and colleagues (Reference Tse, Codling, Jones, Thoms, Liber, Giesy, Wheater and Doig2014:303), however, shows a clear increase in coprostanol during a period of historic sewage discharge to Ross Lake in Manitoba and subsequent decrease following the construction of a wastewater treatment facility. Tse and colleagues (Reference Tse, Codling, Jones, Thoms, Liber, Giesy, Wheater and Doig2014:304) attribute higher coprostanol/cholesterol ratio values in postindustrial sediments to continued sewage input following the installation of the treatment facility—but at levels lower than before. Consequently, the Tse and colleagues study strengthens our interpretation by showing a tight correlation between human activity and fecal stanol concentration in lake sediments. Finally, our fecal stanol record closely tracks Mississippian-period population estimates derived from the archaeological record (Milner Reference Milner2006; Pauketat and Lopinot Reference Pauketat, Lopinot, Pauketat and Emerson1997), supporting the inference that fecal stanols track population change within the Horseshoe Lake watershed (White et al. Reference White, Stevens, Lorenzi, Munoz, Lipo and Schroeder2018).

Skousen and Aiuvalasit further state that we “discount regional archaeological and historical data.” We welcome this comment because it raises a significant question: does a lack of archaeological evidence for an occupation preclude its existence, even in a well-studied area such as Cahokia? Skousen and Aiuvalasit note that “more than 480 km² have been surveyed in the three counties that comprise the American Bottom region,” an archaeological survey coverage of about 10%. The geochemical and paleoenvironmental methods that we use both augment and complement traditional archaeological procedures of reconstructing the past, but they are necessarily focused on the Horseshoe Lake watershed rather than the entire American Bottom region. Giving primacy to archaeological data at the expense of other lines of evidence comes with the risk of overlooking traces of past peoples who did not leave behind significant or recognizable material signatures of their presence.

For example, archaeological evidence exists for the River L'Abbe Mission on Monks Mound, but beyond a few early historic artifacts and features on and near Monks Mound, we are unaware of other archaeological traces of the population of Cahokians for whom the mission was built (Benchley Reference Benchley1981; Mazrim and Esarey Reference Mazrim and Esarey2007:164; Walthall and Benchley Reference Walthall and Benchley1987:4). Furthermore, despite historic documentation of the Illinois having an extensive presence during the winter in the lower Illinois River valley, there is only limited archaeological evidence for their encampments (Mazrim and Esarey Reference Mazrim and Esarey2007:163). Does a lack of archaeological evidence for a village near the River L'Abbe Mission or sites in the lower Illinois River valley mean that the historical records are wrong? Or, do these sites exist, but substantial archaeological evidence for them has not yet been found?

Historical records report the presence of members of the Illinois Confederacy in Illinois Country and allude to the presence of the Cahokia and Tamaroa in the northern American Bottom in the seventeenth century (Mazrim and Esarey Reference Mazrim and Esarey2007). Ceramic types associated with the Cahokia and Tamaroa have not yet been identified (Mazrim and Esarey Reference Mazrim and Esarey2007:162), so it is an untested assumption that they would look like the Danner series mentioned by Skousen and Aiuvalasit, which has been linked with the Illinois (Mazrim and Esarey Reference Mazrim and Esarey2007). We agree with Skousen and Aiuvalasit that cultural groups other than the Illinois Confederacy may have contributed to the Horseshoe Lake fecal stanol record, and we encourage further research in this area. We maintain, however, that the evidence we provide is consistent with a post-Mississippian population increase in the region centered around AD 1650. Following this period, the fecal stanol record parallels historical sources, suggesting an Indigenous presence in the region through much of the eighteenth century despite a lack of archaeological evidence.

Although powerful, the archaeological record is fragile, destructible, and—like all disciplines—subject to problems of interpretation. To reach a more complete understanding of the past, we must consider multiple lines of evidence in addition to the archaeological record, as opposed to giving it primacy over other means.