Sex and Gender: A Complex Issue

The archaeological literature is full of men and women. For decades, buried individuals have been interpreted as male or female based on grave goods, and the development of osteology and genetics have brought in new methods to determine human remains as either female or male. This binary division is, however, problematic. Chromosomal aneuploidies—as well as certain genetic and hormonal conditions—widen the variation in biological sex (Sofaer, Reference Sofaer, Gowland and Knüsel2006: 157; Fausto-Sterling, Reference Fausto-Sterling2020: 74–76), but biology itself offers little in terms of a person's self-identification. Some recent studies suggest that brains produce personality, cognition, and behaviour similarly regardless of chromosomal sex, in opposition to the idea that binary female and male differences exist in the brain (Joel et al., Reference Joel, Berman, Tavor, Wexler, Gaber and Stein2015, Reference Joel, Garcia-Falgueras and Swaab2020).

In the 1970s, it became common to distinguish biological sex from socially and culturally defined gender (e.g. Fausto-Sterling, Reference Fausto-Sterling2020: 14–16 with references). This concept has since been developed further. Butler (Reference Butler1990) has argued that sex and gender are interlinked social constructs: gender is a performance based on the expectations of society, a set of norms associated with a certain kind of body. While social gender can be understood as representing culturally specific expectations of behaviour, sex can be understood as a common agreement on certain chromosomal, hormonal, anatomical, and physical features (Butler, Reference Butler1990; Fausto-Sterling, Reference Fausto-Sterling2012, Reference Fausto-Sterling2020).

The binary division of sexes is arguably rooted in a modern, western mindset, and gendered norms and expectations have varied culturally, geographically, and temporally (e.g. Laqueur, Reference Laqueur1990: 6–12; Comaroff & Comaroff, Reference Comaroff and Comaroff1992: 78; Motschenbacher, Reference Motschenbacher2010: 13). Early medieval societies evidently had their own expectations and norms for female- and male-bodied individuals, but were these norms inflexible and exclusive? And how were they negotiated and displayed in different circumstances? The traditional view emphasizes rigid masculine ideals (e.g. Clover, Reference Clover1993; Hedeager, Reference Hedeager2011), but many recent studies indicate that early medieval societies understood and performed gender in diverse and nuanced ways (e.g. Back Danielsson, Reference Back Danielsson, Alexandersson, Andreeff and Bünz2014; Moen, Reference Moen2019). This encourages us to think about a gender spectrum instead of a binary gender division (Moen, Reference Moen2019: 217).

The complexity of gender is evident in the problem of determining the sex or gender of individuals based on the artefacts recovered from their graves (e.g. Berg & Lie, Reference Berg and Lie1995; Effros, Reference Effros2000; Stratton, Reference Stratton2016). It is unclear how well the grave goods represent the gender roles and identities of the past, and whether these roles should be interpreted from a binary perspective (e.g. Hjørungdal, Reference Hjørungdal1994; Nixon, Reference Nixon, Archer, Fischler and Wyke1994; Conkey & Gero, Reference Conkey and Gero1997; Gilchrist, Reference Gilchrist1999; Sørensen, Reference Sørensen2000; Joyce, Reference Joyce2008; Hedenstierna-Jonson et al., Reference Hedenstierna-Jonson, Kjellström, Zachrisson, Krzewińska, Sobrado and Price2017; Moen, Reference Moen2019; Price et al., Reference Price, Hedenstierna-Jonson, Zachrisson, Kjellström, Storå and Krzewińska2019). Although swords in early medieval graves in northern Europe can be interpreted in many ways, from symbols of power and social status to ancestral heirlooms (Hadley, Reference Hadley, Crawford and Hamerow2008: 273; Gardela, Reference Gardela2013; Härke, Reference Härke2014; Harrison, Reference Harrison, Barrett and Gibbon2015), they are still often associated with masculinity and warrior ideology (see e.g. Jakobsson, Reference Jakobsson1992; Petré, Reference Petré1993; Raninen, Reference Raninen2006: 4; Hadley, Reference Hadley, Crawford and Hamerow2008; Hedeager, Reference Hedeager2011: 118, 124; Harrison, Reference Harrison, Barrett and Gibbon2015; Sayer et al., Reference Sayer, Sebo and Hughies2019: 561). While axes, spearheads, and arrowheads have occasionally been buried with female-bodied people, and may appear in graves together with feminine jewellery (e.g. Mägi, Reference Mägi2002: 77–79; Gardela, Reference Gardela2013; Moisio, Reference Moisio2016; Lund & Moen, Reference Lund and Moen2019), swords are rare items in graves of female-bodied individuals (Raninen, Reference Raninen and Mäntylä2005: 229; Wessman, Reference Wessman2010: 102; McLeod, Reference McLeod2011: 343; Raninen & Wessman, Reference Raninen, Wessman, Haggrén, Halinen, Lavento, Raninen and Wessman2016: 319–20). An interesting aspect of the graves containing osteologically determined females and swords is that they often lack jewellery and other feminine accessories (Simniškytė, Reference Simniškytė and Bliujienė2007; Price et al., Reference Price, Hedenstierna-Jonson, Zachrisson, Kjellström, Storå and Krzewińska2019). This is seemingly in line with the idea that the Scandinavian gender system accepted masculinity as the only normative gender and allowed only some females to obtain masculine gender in certain circumstances (Clover, Reference Clover1993). Weapons are often considered non-normative items in women's graves, and the emphasized masculine symbolism of swords is the reason why graves containing both ‘typically’ feminine and masculine items are often considered problematic. The combined presence of weapons and jewellery in graves easily leads to interpretations of mixed gender double burials, even if the context indicates only a single skeleton in the grave (see Nylén, Reference Nylén, Sarvas and Siiriäinen1973: 165; Taavitsainen, Reference Taavitsainen1990: 91; McLeod, Reference McLeod2011: 343; Androshchuk, Reference Androshchuk2018). However, as Moen (Reference Moen2019) has pointed out, the graves may not tell us about the gender systems of the past per se, but rather about the assumptions of the modern people making the interpretations.

The Context of the Suontaka Grave

The Suontaka grave is famous for its bronze-hilted sword (National Museum of Finland, hereafter NM, accession number 17777:1), decorated with Urnes style ornaments (Keskitalo, Reference Keskitalo1969b) (Figure 2A). The other finds include a less-discussed, hiltless sword blade (NM 17777:2) with silver inlays, a sheathed knife (NM 17777:3), two oval brooches (NM 17777:4–5), a twin-spiral chain-bearer (NM 17777:6), a small penannular brooch (NM 17777:7), and a sickle (NM 17777:8) (Figure 2B–G). Of these items, the swords are traditionally associated with males and jewellery with females. In Finland, sickles are common in both feminine and masculine graves, although they are slightly more frequently associated with females than with males (Lehtosalo-Hilander, Reference Lehtosalo-Hilander1982: 54–55). The typological dating of the artefacts indicates that the burial took place in the late eleventh or early twelfth century ad (Taavitsainen, Reference Taavitsainen1990: 91), which coincides with the Crusade period (ad 1025–1150) in the western Finnish chronology. Radiocarbon dating supports this, as a femur fragment from the grave gave a date of 921 ± 17 bp (Hela-4566), i.e. ad 1040–1174 at 95.4 per cent; (date modelled in OxCal v4.4.2 (Bronk Ramsey, Reference Bronk Ramsey2017), using the IntCal13 atmospheric curve (Reimer et al., Reference Reimer, Austin, Bard, Bayliss, Blackwell and Bronk Ramsey2020)).

Figure 2.

The objects found in the Suontaka grave. A: bronze-hilted sword (NM 17777:1); B: hiltless sword (NM 17777:2) with silver inlays (inset); C: two oval brooches with textile fragments (NM 17777:4–5); D: twin-spiral chain-bearer (NM 17777:6); E: sheathed knife (NM 17777:3); F: penannular brooch (NM 17777:7); G: sickle (NM 17777:8). A: © Finnish Heritage Agency.

The original excavation report from 1969 describes the Suontaka grave and the finds in detail (Keskitalo, Reference Keskitalo1969a). The excavation was carried out in late October 1968 after the bronze-hilted sword was found by water pipeline workers. Oiva Keskitalo, who excavated the site, noticed a dark-coloured area—the grave cut—in the water pipe trench. According to his description, the dark area was the grave pit's upper right corner, where the bronze-hilted sword had been located (Keskitalo, Reference Keskitalo1969a). Although Keskitalo does not mention the exact distance between the sword and the other finds, his photographs indicate that he had to remove some of the dark soil before reaching the skeletal remains and the other artefacts. This suggests that the bronze-hilted sword may have been located in the grave fill, and therefore it is unlikely to be directly associated with the burial assemblage. No other graves are known from the vicinity, and it is unknown whether the burial represents a solitary burial or whether it is part of a larger cemetery.

In the literature, the Suontaka burial has been called ‘partly destroyed’ because of the bronze-hilted sword find and the manner of its discovery. Jussi-Pekka Taavitsainen, who has been critical of its interpretation as a female weapon burial, emphasizes the unreliability of the find context by saying that the grave was excavated in ‘difficult conditions’ (Taavitsainen, Reference Taavitsainen1990: 91). Although the excavation was carried out in late autumn when the temperature had dropped to -10°C, Keskitalo states in his original report (Reference Keskitalo1969a) that only the surface of the ground was frozen. This suggests that the conditions at the bottom of the grave pit (70–90 cm from the topsoil) where the skeleton was lying, were not significantly different from those of summertime excavations.Footnote ² Therefore, the field observations should be considered reliable. The position and the shape of the skeleton were observed and documented: the tibias, femora, pelvis, elbow joints, ribs, and skull were in situ (Figure 3A). Although observable, the bones were almost fully decomposed and consisted of a soft mass, and only fragments of two femora (NM 17777:9) could be lifted from the ground.

Figure 3.

A: plan of the Suontaka burial. ‘Täckdike’ marks the water pipe trench which led to the discovery of the grave (Keskitalo, Reference Keskitalo1969b: 85). B: artist's reconstruction of the burial, showing the position of the objects on the body. A reproduced by permission of Finnish Heritage Agency. B: drawing by Veronika Paschenko.

As mentioned, it is rare in a Nordic context to find a sword in a grave with several artefacts with feminine gender association. Because of the unusual combination of objects, it has been suggested that there must have been two individuals in the grave (Nylén, Reference Nylén, Sarvas and Siiriäinen1973: 165; Taavitsainen, Reference Taavitsainen1990: 91). Keskitalo was himself puzzled by the fact that the skeleton was equipped with two brooches (usually considered part of feminine dress, see Lehtosalo-Hilander, Reference Lehtosalo-Hilander1984) and a sword (traditionally associated with male burials), and he tried to find evidence of more than one individual in the grave without success. Keskitalo (Reference Keskitalo1969a, Reference Keskitalo1969b) observed that the burial pit had an even floor, where the corpse had been placed in a supine position without a coffin. The size of the pit was suitable for only one corpse. If there had been another, it would have been on top of the first. Although placing several bodies in the same grave on top of each other may influence the decomposition process by accelerating the speed of decay in the uppermost bodies (Troutman et al., Reference Troutman, Moffatt and Simmons2014), it is implausible that the Suontaka grave was initially made for two people. No different colouring of the soil indicated another decomposition layer, and no other artefacts were observed in the upper parts of the grave. Even if there had been another corpse above the documented skeleton, its only accompanying item would possibly have been the bronze-hilted sword.

Apart from the bronze-hilted sword, all the other objects were directly connected with the skeletal remains (Figure 3A). The hiltless sword blade was located on the left side of the pelvis and the sheathed knife on top of the sword's blade. Two oval brooches with woollen textile fragments were located below the shoulders, indicating that the individual had been buried in a typical feminine costume of the era (see Lehtosalo-Hilander, Reference Lehtosalo-Hilander1984). The twin-spiral chain-bearer was located on the chest, between the brooches. Based on its position, and on the fact that no chain loops were found from the grave, the object may have been used as a pendant and not as a chain-bearer, the latter being a distinctive artefact in typical female graves (Figure 3B). The penannular brooch was found approximately at the waist level, and the sickle had been placed on the individual's upper chest. Items placed on the chest or stomach might move slightly downwards during the decomposition of soft tissues, but as the burial did not contain a coffin, the sandy soil constantly replaced the decaying tissues (Duday, Reference Duday2009). The displacement of objects and bones is therefore not as extensive as when decomposing in a void created by a coffin or other structure. Because of this, it seems clear that all the items—except the bronze-hilted sword—were placed directly on the body. Hence, these artefacts can be associated with the buried individual, and not another corpse that would have completely decomposed.

Fibres and Animal Hairs in the Grave

Separate soil samples were not taken from the grave during the excavation, but the femur fragments were lifted from the ground with a small amount of soil. This soil represents the original material in the grave, in direct contact with the buried body. The soil was analysed for possible animal hairs and textile fibres (see Supplementary Material 1 for details and methods).

In total, twenty-three mammalian hairs and three bird feather fragments were extracted from the sample (Supplementary Material 1, Table S1). Most of the hairs were very short, from 0.2 to 2 mm long, and for this reason, only part of them could be identified to species or family level. Some 60 per cent of the hairs (n = 14), were identified as belonging to sheep (Ovis aries) fine hairs, i.e. wool (Figure 4C). Most were natural white (n = 6) or brown (n = 3), and single hairs were coloured blue (n = 2) or blueish/green (n = 3). Seven non-medullated, colourless hairs were not identifiable. It is possible that at least some of these might have been fine hairs from sheep, too. In addition to wool fibres, a fine hair of a fur-bearing animal, probably red fox (Vulpes vulpes) or a Mustelid, and a fine hair of a rabbit or hare (Leporidae) were discovered (Figure 4A–B). The latter was exceptional because it was purple in colour, although that might be of secondary origin. Besides mammalian hairs, three minuscule (0.2–0.5 μm) bird feather fragments, so-called barbules, were discovered (Figure 4D). These fragments showed no diagnostic features for species identification.

Figure 4.

Examples of identified animal hairs from the soil sample. A: Leporidae (see Supplementary Material 1, Table S1, K3); B: Vulpes vulpes or Mustelidae (K20); C: Ovis aries (K13); D: Aves (K19).

Determining the Chromosomal Sex

As bone material from the Suontaka grave consists of only two highly degraded femur fragments, an osteological analysis could not be carried out. Ancient DNA (aDNA) analyses, on the other hand, may be used to infer the chromosomal sex of an individual even from low quantities of skeletal material (Reich et al., Reference Reich, Green, Kircher, Krause, Patterson and Durand2010). We extracted and sequenced aDNA from one of the femur fragments in the archaeogenetics laboratory of the Max Planck Institute for the Science of Human History in Jena, Germany (see Supplementary Material 2 for a detailed description of the laboratory analysis protocols). Unfortunately, the sample gave a very low yield of endogenous human DNA: even after a capture procedure to enrich for human DNA, the data contained only 106,781 sequence reads mapping to the human genome (of a total of 18,250,176 overall reads), 8329 of which had a mapping quality above 30. Of these, 2534 also showed post-mortem damage (PMD) scores above 0 from PMDtools (Skoglund et al., Reference Skoglund, Northoff, Shunkov, Derevianko, Pääbo, Krause and Jakobsson2014) (Supplementary Material 2, Table S2). This paucity of data seriously limited the range of aDNA analyses that could be conducted, including those for data authentication (see Supplementary Material 2). Therefore, we did not extend our genetic analyses beyond sex determination, as that is among the analysis types that need the least amount of data.

Existing methods for chromosomal sex determination (Skoglund et al., Reference Skoglund, Storå, Götherström and Jakobsson2013; Fu et al., Reference Fu, Posth, Hajdinjak, Petr, Mallick and Fernandes2016; Mittnik et al., Reference Mittnik, Wand, Svoboda and Krause2016) lack power for data as sparse as this, but their results suggested that the Suontaka individual's X-chromosomal and Y-chromosomal read counts fit neither those expected for XX (female) nor for XY (male) individuals. We therefore developed a novel approach to estimate the chromosomal sex of the individual, where we downsampled sequencing reads from individuals of known genetic sex to the number of reads observed in the Suontaka individual (n = 8329) to model four possible scenarios that could have produced the observed data: XX, XY, a contaminated sample with a mix of reads from XX and XY individuals, and an aneuploidic karyotype XXY (male with Klinefelter syndrome). The details and assumptions of the model are described in the Supplementary Material 2.

Despite the extremely low sequencing coverage, we found overwhelming evidence that the genetic data of the Suontaka individual most closely resemble an XXY karyotype: our model classified the Suontaka individual as XXY at a 99.75 per cent probability, as contaminated with a 0.25 per cent probability, and as XX or XY with a very low probability (less than 10^-6 in either case). In a subset of data that was enriched for plausibly ancient-looking reads (n = 2534), the corresponding probabilities were 99.96 per cent for XXY, 0.04 per cent for contamination, and again negligible for XX and XY; it therefore seems that the XXY signal in the data is not driven by potential modern contaminating DNA. Even when we take into account the low population frequency of the XXY karyotype, the XX and XY scenarios remain extremely unlikely, and—unless we assume a considerably high prior probability of contamination—the Suontaka individual's karyotype is still most likely to be XXY.

The condition in which males are born with one or more extra X chromosomes is known as Klinefelter syndrome. With its prevalence of 1 in 576 male births, XXY is the most common sex-chromosomal aneuploidy in humans (Nielsen & Wohlert, Reference Nielsen and Wohlert1991). The clinical signs of karyotype XXY vary from very subtle and unnoticeable to apparent differences in physical features. The anatomical appearance of XXY individuals is male, and some of them never even notice that they have the condition. In some cases, the clinical signs are stronger: XXY males can be infertile and have hypospadias (the opening of the urethra is on the underside of the penis), small phallus and testicles, and gynecomastia (breast growth). Testosterone deficiency may cause delayed or incomplete pubertal development. Sometimes effects on physical and cognitive development are reported and, according to some studies, modern XXY males may consider themselves more sensitive and unassertive than others (Visootsak & Graham, Reference Visootsak and Graham2006). Interviews also suggest gender-related insecurities stemming from the XXY males feeling physically more feminine than other males (Berch & McCauley, Reference Berch, McCauley and Holmes1990). Because the modern XXY males may compare their experiences to modern expectations of sex and gender, it is difficult to say how the physical and possibly psychological aspects would have been understood and displayed in eleventh–twelfth-century Finland.