Direct isotopic evidence for human millet consumption in the Middle Mumun period: Implication and importance of millets in early agriculture on the Korean Peninsula

Highlights

• Bronze Age Mumun individuals consumed C₄ plants (foxtail and broomcorn millet.

• Results counter the current emphasis of rice farming during the Mumun period.

• Findings provide a reevaluation of millets in early agriculture on Korean Peninsula.

Abstract

It is generally believed that early agriculture on the Korean Peninsula was established during the Mumun period (1500–100 BC). While previous studies on agriculture in prehistoric Korea have relied on cultivated plant remains from archaeological sites, only a few isotopic studies have been conducted on Mumun individuals due to poor bone preservation during this period. Here, we measured bulk carbon (δ¹³C) and nitrogen (δ¹⁵N) stable isotope ratios as well as individual amino acid δ¹³C results (δ¹³C_AA) of collagen from human (n = 7) and animal (n = 4) bones from three Mumun sites (Hwangsok-ri, Jungdo, Maedun Cave) in the central inland portion of South Korea. The aims of this study were to explore the contribution of plant foods to the human diet and to examine the type and extent of agriculture in the Mumun period. In contrast to the surrounding C₃ vegetation, all the Mumun humans in this study had significantly ¹³C-enriched results, evidence for the consumption of C₄ plants (foxtail and broomcorn millet). The δ¹³C_AA data show that there was no consumption of freshwater or marine resources in the diet of the Mumun. These data indicate that C₄ plants (millets) were the main dietary sources in central inland South Korea and that millet agriculture was fully established during the Middle Mumun period. This finding highlights the importance of millet cultivation during the Mumun period and provides a revaluation for the significance of millets in the development of early agriculture on the Korean Peninsula.

Introduction

The origins, cultivation, and consumption of domesticated plants on the Korean Peninsula in prehistory is debated due to the paucity of direct dietary evidence (Ahn, 2005; 2006; 2010; Bale, 2001; Crawford and Lee 2003; 2011; Kim, 2015; Kwak et al., 2017; Kim and Park, 2020). Most previous studies focused on the development of early agriculture on the Korean Peninsula were based on the analyses of plant and animal remains from excavated sites (Lee, 2001, 2003, Lee, 2011; Crawford and Lee, 2003). Based on the archaeobotanical evidence, it is assumed that early agriculture was intensified during the Bronze Age in Korean prehistory or the Mumun period (1500–100 BC) (Ahn, 2010; 2011; Kwak et al., 2017). Although the earliest evidence for cultivated plants is reported from the Middle Chulmun period (3500–2000 BC), it is believed that domesticated crops started to be important from the Early Mumun period (1500–800 BC) (Lee, 2011; Bale, 2001; Lee and Bale, 2016). The Mumun people cultivated various domesticated plants such as rice (Oryza sativa), foxtail millet (Setaria italica), broomcorn millet (Panicum miliaceum), six-rowed barley (Hordeum vulgare), naked bread wheat (Triticum aestivum), soybean (Glycine max), and adzuki bean (Vigna angularis) (Lee, 2003; Ahn, 2010). Despite abundant evidence for the cultivation of multiple crops during the Mumun period, most previous studies were mainly focused on the importance of rice agriculture (Kim, B–C., 2005; 2006; Cho, 2008; Ahn, 2010; Kim, M., 2015). Along with rice, other crops also are believed to be important foods for the Mumun people (Ahn, 2010; Lee, 2011). However, previous archaeobotanical research on domesticated crops has been limited in its ability to directly determine the relative contribution of these crops to the whole diet. In addition, recent studies on organic residues in pottery also have provided evidence of the presence rather than the dietary significance, of different food resources in the Mumun diets (Kwak and Marwick, 2015; Shoda et al., 2017; Kwak et al., 2017). Until now, little research has been conducted on the direct dietary significance of each cultivated crop to the whole Mumun diet. Hence, it is necessary to investigate the relative contribution of each crop (e.g. rice vs. millet) in terms of the Mumun diet in order to better understand the nature and type of agriculture that the Mumun engaged in on the Korean Peninsula.

While archaeobotanical analysis of cultivated cereals provides valuable information about the dietary resources of past populations (Ahn, 2010; Kim, 2015), the presence of plant remains in excavated sites does not imply direct human consumption of these plant remains because they can be encountered as artifacts or as natural deposits in archaeological sites (Kusaka et al., 2010). In order to bridge these disparities between the preservation and direct consumption of plant remains, stable isotope ratio analysis of human bone collagen is applied here. This approach is an established method in bioarchaeology used to reconstruct direct dietary patterns of ancient individuals or population (Lee-Thorp, 2008; Richards, 2020). Isotopic measurements of bone collagen provide direct evidence of the long-term dietary record of an individual related to the average foods consumed over their lifetime (Katzenberg, 2000; Richards, 2020). Specifically, carbon stable isotope ratios (δ¹³C) distinguish diets based on the consumption of plants that use different photosynthetic pathways (C₃ vs. C₄ plants) (Wang et al., 2019) and between the consumption of terrestrial and marine foods (Richards and Hedges, 1999; Lee-Thorp, 2008). In Asia, the main C₄ domestic plants, broomcorn and foxtail millet, are significantly ¹³C-enriched, which results in measurable differences in the bone collagen of consumers of these foods (Chen et al., 2016, Lu et al., 2009; Dai et al., 2016; Ma et al., 2016a, 2016b, 2016c; Wang et al., 2016, 2018, 2019).

Bulk nitrogen stable isotope ratio measurements (δ¹⁵N) provide information about the trophic level of a consumer in a food chain, with human and animal bone collagen ¹⁵N-enriched by ~3–5‰ compared to dietary sources (Minagawa and Wada, 1984; Bocherens and Drucker, 2003; Hedges and Reynard, 2007). Thus, human δ¹⁵N values higher than associated faunal δ¹⁵N values are interpreted as indicating a diet with significant quantities of these animals, whereas δ¹⁵N values close to the associated herbivore values are interpreted as indicating a diet based mainly on plant proteins (Lee-Thorp, 2008).

In addition, compound specific stable isotope ratio analysis of individual amino acids (CSIA-AA) is used to overcome limitations associated with bulk carbon and nitrogen isotopic measurements and provide more detailed dietary reconstructions (Hare et al., 1991; Fogel and Tuross, 2003; Smith et al., 2009; Jaouen et al., 2019). In particular, carbon amino acid measurements (δ¹³C_AA) can better distinguish between C₃ vs. freshwater consumers as well as C₄ vs. marine consumers in environmental settings with diverse dietary resources (Corr et al., 2005; Choy et al., 2010a; Honch et al., 2012; Webb et al., 2015, 2018; Ma et al., 2021). Further, nitrogen amino acid measurements (δ¹⁵N_AA) can provide more accurate trophic level determinations without direct comparisons of associated faunal specimens (Fuller and Petzke, 2017; Ohkouchi et al., 2017; Jaouen et al., 2019). An in-depth discussion of bulk and CSIA-AA is beyond the scope of this work, and the following reviews should be consulted for additional details (Lee-Thorp, 2008; Reitsema, 2015; McMahon and Newsome, 2019; Richards, 2020).

In Korean archaeology, the application of bulk and CSIA-AA analysis to palaeodiet reconstruction has provided important information about subsistence practices in prehistoric populations (Choy and Richards, 2009, 2010b; Choy et al., 2010a, 2012; Shin and Lee, 2009; Kim, H–S., 2010). In particular, isotopic results revealed that the Chulmun (8000–1500 BC) people were mainly dependent on marine food resources before the introduction of domesticated crops (Choy et al., 2012). However, after the arrival of domesticated crops on the Korean Peninsula, during the Mumun period (1500–100 BC) there are few applications of stable isotope analysis to human bone collagen largely due to the lack of human remains from this time. Most previous studies on dietary patterns during the Mumun period are based on the analysis of charred plant remains (Lee, 2001; Lee, 2003; Crawford and Lee, 2003; Ahn, 2010; Kim, M., 2015). In this study, we were fortunate to obtain access to human (n = 7) and animal bones (n = 4) from three Mumun sites (Hwangsok-ri, Jungdo, Maedun Cave), which are all located in the inland portion of central South Korea (Fig. 1). Bulk and δ¹³C_AA analyses were applied to these specimens to directly reconstruct the dietary patterns of the Mumun people. Direct radiocarbon measurements were also conducted on three of these humans to firmly anchor these isotopic measurements in time. In addition, the contribution of each food source to the entire human diet is investigated and the importance of agriculture crops in terms of human consumption patterns, is examined during the Mumun period in South Korea.