Introduction

Patrilocality, suggested by both strontium isotope (87Sr/86Sr) analysis (Bentley et al., 2003, 2012) and aDNA studies (Rasteiro & Chikhi, 2013; Rasteiro et al., 2012), has become the accepted and dominant kinship model for the Early Neolithic in Central Europe (Linearbandkeramik – hereafter LBK: c. 5500–4900 cal BC). Such exogamic practices are often implicitly linked to strongly gender-differentiated demographic and dietary patterns (Bickle & Whittle, 2013; Masclans et al., 2021a; Oelze et al., 2011), sexual division of labour (Masclans et al., 2021a) and interpersonal violence (Hedges et al., 2013), leading to assumptions about female status and male dominance (Jeunesse, 1997; Augereau, 2018).

The western Carpathian Basin is home to some of the most studied Early Neolithic cemeteries, such as Vedrovice (Czech Republic) and Nitra (Slovakia). They have been included in several research projects involving Early Neolithic male and female mobility patterns through 87Sr/86Sr isotopic analysis (Smrčka, 2005; Richards et al., 2008; Bickle & Whittle, 2013). These studies suggest that, while most male individuals remained within the same area all their lives, female mobility was more variable, which has been interpreted in terms of exogamic patrilocal practices. Polygyny has also been recently suggested, given the demographic skew towards females at the two cemeteries (Hrnčíř et al., 2020a, b). However, alternative interpretations of the same data proposed that bilocality and bilateral descent patterns were present at Vedrovice, and local marriages with groups with redundant resources occurred at Nitra (Ensor, 2021). Kinship need not explain all mobility data and Hofmann (2020) hypothesised migration was a social strategy to improve individual status.

Indeed, females from Vedrovice displayed greater variability in the 87Sr/86Sr isotopic values of their second molar (M2, mineralised between 1 and 8 years), than in their first (M1, mineralized during the last month of gestation up to 2 years) and third molar (M3, mineralised between 7 and 17 years) (Bickle & Whittle, 2013; Masclans et al., 2020), which implies further complexity.

A possible correlation between the mobility patterns, the δ15N values and the grave-good characterization among cemetery individuals (Vedrovice-Široká u Lesa) have been suggested. Whittle et al. (2013) and Bentley et al. (2012) found that males buried with adzes were consistently related to local loess areas, while Masclans et al., (2021a, 2021b) pointed out that local skeletons had higher δ15N values than the non-local individuals (related either to a diet richer in protein, or to the consumption of animals that grazed on fields with different degrees of fertilisation), and that the non-locals (who were predominantly females) were less probably related with stone and bone tools and most likely buried with pottery vessels and Spondylus ornaments.

These mobility patterns have been predominantly observed in cemeteries, which account for the largest percentage of known LBK burials in the area. However, other kinds of funerary contexts in south-eastern Moravia have also been studied for 87Sr/86Sr analysis, including burials alongside houses, settlement burials (Vedrovice-Sidliště, Brno-Starý Lískovec and some of the Těšetice-Kyjovice graves) and small burial clusters within settlements (including Vedrovice-Za Dvorem and some of the Těšetice-Kyjovice burials) (Bickle et al., 2014; Whittle et al., 2013). These contexts generally contain graves similar to those found at cemeteries—oval pits with individual inhumations—but present more variability in body positions and orientations, suggesting that individuals buried in cemeteries more closely followed a normative LBK burial rite than those buried outside them (Bickle et al., 2014; Hofmann, 2009). Furthermore, it has been suggested that settlements and cemeteries generally present different demographic patterns, with females and non-adults generally found in higher numbers in settlements and considered to have low status because of their lower number of grave goods (as summarised by Hofmann & Orschiedt, 2015). However, recent reanalysis suggests that the belief that more females than males were buried at settlements may have been overemphasised (Bickle, 2020).

As for 87Sr/86Sr analysis, settlement graves and small clusters of burials have been considered to fall within isotopic local ranges, while a greater value variability has been observed in cemeteries (Bickle et al., 2014; Whittle et al., 2013). Until very recently, the settlement funerary contexts were not dated through radiocarbon dating, preventing rigorous comparisons with cemetery data. Recent studies have provided a robust knowledge of the temporality of the contexts included in this research, confirming that they overlapped in time (Masclans et al., in preparation). Specifically, Bayesian modelling determined that all the dated burials were continuous in time and distributed in three overlapping phases (Masclans et al., in preparation; Fig. 1, see Online Resource SI 2: Table 1), indicating that the funerary practices are contemporaneous in time. This fact implies that any differences in the 87Sr/86Sr and δ18O isotopic data can be considered as arising in different concurrent lifeways, rather than representing change over time, raising new and interesting interpretative possibilities.

Fig. 1
figure 1

Overlapping phases Bayesian modelling (Amodel 108.5 and Aoverall 107.8) of the analysed sites using OxCal v.4.4 software (Bronk Ramsey, 2009) and the IntCal20 calibration curve (Reimer et al., 2020)

Full size image

This data has had an impact on our knowledge and understanding of LBK communities and their lifeways. However, the empirical evidence to support this hypothesis is presently constrained for five main reasons. Firstly, there is an over-representation of samples from cemeteries over settlement burials. A total of 68 87Sr/86Sr samples from 63 individuals are available in the cemetery of Vedrovice-Široká u Lesa (72.41% of the inhumed sampled), while just 28 samples on 26 individuals were performed in the settlement burials of Vedrovice-Sidliště, Brno-Starý Lískovec and Těšetice-Kyjovice, and the Vedrovice-Za Dvorem small cluster of burials (54.16% of the inhumed sampled). This fact may exaggerate the variability of cemeteries, suggesting that they represent a more heterogeneous population in their isotopic values than the settlements. The different demographics of settlements and cemeteries may also add to these challenges, as the greater presence of certain sex and age ranges can influence isotopic variability. Secondly, most of the available samples were taken from M1 (54.41% of the samples) and M2 (58.82%) molars, while M3 is underrepresented (30.88%), which makes it difficult to evaluate whether indeed age played a determining role in the mobility patterns of these communities.

The characterisation of the local isotopic 87Sr/86Sr footprint is also necessary. Previous works estimated the local isotopic footprint by means of human femur samples (Smrčka et al., 2005), calculating averages within the isotopic ranges of human enamel, or by using the range of values for children and juveniles, assuming that children were more likely to be local because they had less time to migrate than adults in their lifetime (Bickle & Whittle, 2013; Hrnčíř & Laffoon, 2019; Masclans, et al., 2021a). However, a more accurate review of the different geological areas surrounding the site’s isotopic values is essential to test the population’s mobility patterns. Considering the limitations of 87Sr/86Sr analysis for the exploration mobility within large areas displaying the same isotopic ratio (Montgomery, 2010; Price et al., 2002), it can be useful to compare strontium data with other mobility proxies, such as δ18O isotopic ratios (Bentley & Knipper, 2005a, 2005b). This complementary approach would make it possible to determine whether or not these movements took place in areas with different precipitation indices or water resources (Pederzani & Britton, 2019).

Finally, the abovementioned models do not fully take account of the implications of the biased demographic pyramid, in both cemeteries and settlement sites, when interpreting mobility patterns. This is especially the case in the male sample, which presents clear age gaps between sites. A thorough review of all the published anthropological data is still necessary to reinterpret the consistency of the sex and age ratios in each of the different kinds of funerary contexts.

Following the different gaps that limit the explanation and interpretation of the mobility patterns of these communities, the goals of this paper are as follows: (1) to contribute to the characterisation of the south-eastern Moravian 87Sr/86Sr isotopic footprint; (2) to re-evaluate the implications of the population demography in each of the analysed sites for mobility by reviewing all the published anthropological data of the analysed sites; (3) to determine the individual's mobility sequence through their lives; (4) to assess whether the mobility sequences changed according to the type of funerary pattern (cemeteries/settlement graves); (5) to correlate 87Sr/86Sr mobility data with δ18O values in order to test whether the two mobility indicators match; and (6) to re-evaluate the correlation between the 87Sr/86Sr and δ18O isotopic values, the variability in the inhumed person’s grave goods, and diet (δ15N and δ13C) in the light of the new data.

Materials and Methods

To achieve our goals, a total of 5 funerary contexts have been analysed, namely Brno-Starý Lískovec, Těšetice-Kyjovice, Vedrovice-Sidliště, Vedrovice-Široká u Lesa, and Vedrovice-Za Dvorem. Vedrovice (Znojmo District, Fig. 2) is located in Southern Moravia at the base of the Bohemian Massif, between the rivers Dyje (south) and Jihlava (north). It comprises three different burial areas: the settlement (Sidliště), a cemetery (Široká u Lesa) and a small grave cluster (Za Dvorem). On the basis of site documentation and osteological analyses (Podborský, 2002; Ondruš, 1972; Smrčka et al., 2005; Dočkalová & Čižmář, 2007, 2008), and discarding non-reliable grave contexts, 86 individuals distributed in 84 graves were counted at Široká u Lesa, 10 burials at Sidliště, and 13 at Za Dvorem, making a total of 109 individuals (SI 1, Table 2).

Fig. 2
figure 2

Map displaying the sites included in the analysis: Těšetice-Kyjovice (1), Vedrovice Široká u Lesa, Sidliště and Za Dvorem (2) and Brno-Starý Lískovec (3). Map source: OpenStreetMap (Open Database License)

Full size image

Brno-Starý Lískovec (Brno District) is situated on the Bohunice plateau, on the southern outskirts of the city of Brno. It is an enclosed multi-phase settlement with a total of 12 settlement graves (Berkovec, 2004; Přichystal, 2008; Dočkalová & Čižmář, 2007, 2008) (Online Resource SI 1: Table 2). Finally, Těšetice-Kyjovice (Znojmo District) corresponds to a settlement area with a small cluster of 8 graves in its north-eastern part (including one double burial) and 4 burials spread among the dwellings (Online Resource SI 1: Table 2) (Dočkalová, 2006; Dočkalová & Čižmář, 2007, 2008). The site is located between the Bohemian Massif and the Carpathian area, on the eastern edge of the Krumlovian Forest.

A total of 35 new samples were performed for 87Sr/86Sr and δ18O, including both human (26 samples from 13 individuals) and faunal (9 samples from Bos, Ovis and Sus) enamel (Table 1, see Online Resource SI 1: Table 1 for details). Eight samples came from Brno-Starý Lískovec, 10 from Těšetice-Kyjovice, 2 from Vedrovice-Sidliště, 14 from Vedrovice-Za Dvorem and 1 from Vedrovice-Siroká u Lesa.

Table 1 Sex classification of individuals (after Dočkalová & Čižmář 2007, 2008; Lillie, 2008) according to the different sites and the number of individuals sampled for 87Sr/86Sr analysis
Full size table

Previous 87Sr/86Sr performed by Richards et al. (2008), Bickle and Whittle (2013) and Smrčka et al. (2005) were considered during the sample selection, prioritising individuals that had already been sampled with the aim of comparing values between different teeth. Since each molar mineralizes at different times in the first years of life (from childhood to early adolescence), analysing more than one molar per individual allowed us to determine at what age individuals moved.

A total of 119 human enamel samples (including existing and new measurements) from a total of 90 individuals have been used in this study (41 females, 29 males, 1 non-determined adult and 19 non-adults; see Table 1, and SI 1: Table 2 for details – new values are marked in bold). This corresponds to a total of 41 M1, 48 M2, 29 M3 and 1 PM.

The 24 available 87Sr/86Sr values on femur samples from Vedrovice-Sídliste (6), Vedrovice-Za Dvorem (1) and Vedrovice-Siroká u Lesa (1) (Smrčka et al., 2005) were used to characterises the local isotopic footprint (Lacroix, 1971; Price et al., 2002), together with 9 values on fauna samples performed in the framework of this project (Online Resource SI 1: Table 2). Furthermore, a database including 87Sr/86Sr values from Moravian human, geological and animal samples was performed (see Online Resource SI 1: Table 3) and grouped according to the geographical and geological formations where the samples were taken (excluding those coming from migratory animals) (Fig. 3) (see Online Resource SI 1: Table 4). These formations included the Tertiary and Quaternary soils from the Carpathian Foredeep, the Ždánický les mountains (corresponding to the beginning of the Carpathians), the Quaternary soils east of the Ždánický les, the Bohemian Massif Paleozoic and the Bohemian Massif Proterozoic (Chytrý, 2012). To work with as much accuracy as possible, only the values from geological samples, snails and Sus Scrofa (the domestic animal more likely to be less mobile) were ultimately considered (see Online Resource SI 2: Fig. 1).

Fig. 3
figure 3

Available local isotopic signatures of the studied area, based on geological and archaeological/anthropological fauna and malacofauna samples published so far from the following sites: (1) Dolní Vestonice, (2) Držovice, (3) Hluboké Mašůvky, (4) Holubice, (5) Kralice na Hané, (6) Krumlovský les, (7) Kůlna Cave, (8) Mašovice u Znojma, (9) Mikulčice-Valy, (10) Mladeč Cave, (11) Moravská Nová Ves, (12) Nová Ves u Oslavan, (13) Opava, (14) Pohořelice-Šumice, (15) Předmostí-Dluhonice, 16) Rybníky, (17) Seloutky, (18) Slatinky, Močílky, (19) Střelice, (20) Trstěnice, (21) Vyškov, (22) Žadovice, (23) Želešice u Brna and (24) Cezavy u Blučiny. To check the isotopic signatures for each of the sites see: SI 1: Table 3. Map source: Česká geologická služba

Full size image

The chemical pre-treatment of the 87Sr/86Sr and δ18O isotopic samples was conducted at the isotope laboratory of the biogeology group at the Department of Geosciences of the University of Tübingen in accordance with established procedures (Müller-Sohnius, 2007, Bocherens et al. 2016) (for details see Online Resource SI 3). The isotopic measurements were carried out at the Curt-Engelhorn-Zentrum Archäometrie gGmbH in Mannheim, Germany.

The statistical analysis has been performed using R software (version 4.0.1; R Core Team, 2013) in the case of the Multiple Correspondence Analysis, together with the R packages FactoMineR and factoextra (Hervé et al., 2010; Husson et al., 2017) and ggplot2 (Wickham, 2016). χ2, F, T, Kolmogorov–Smirnov, and Shapiro Wilk tests were also used by means of PAST (version 3.0; Hammer et al., 2001).

In all the sampled sites the graves are those characteristic of LBK burial practices: generally oval, and dug directly in the ground, where the bodies were inhumed individually. The grave goods generally consist of ornaments (Spondylus beads, pendants and medallions, snail shells, marble beads and animal teeth), stone adzes, flint blades and fragments, pottery vessels, pebbles possibly used as utensils, bone tools, antler items, ochre blocks, ochre powder, as well as a group of objects generally referred to as ‘grinding tools’, whose precise use is yet to be determined (Podborský, 2002, see Online Resource SI 1: Table 2).

Results

The Anthropological Record

The revision of the anthropological record (Dočkalová & Čižmář, 2007, 2008; Lillie, 2008) found that female-sexed skeletons outnumber males at both the Vedrovice-Široká u Lesa cemetery and Vedrovice-Za Dvorem cluster, while the situation is reversed at the settlement graves, where males and non-adults dominate (Table 1).

If the age of individuals at death is considered (Table 2), opposite trends can also be observed between sites, especially among males. For instance, at Vedrovice-Široká u Lesa a smaller number of adult mature males (5 individuals) are found in contrast to mature adult females (14 individuals). At Vedrovice-Sidliště, the only buried adults are mature males (2 individuals), at Vedrovice-Za Dvorem males are only young adults (2 individuals) and at Těšetice-Kyjovice just one juvenile and two young adults (3 individuals). The only burial place with an equal mixture of male ages is Brno-Starý Lískovec, where 2 adult matures, 2 juveniles, 1 young adult and 1 advanced age adult were documented.

Table 2 Total number of individuals per site according to sex and age. Adult individuals whose sex was not determined were not recorded in this table (for more detailed information see SI 1, Table 2). Age classification criteria follow White et al. (2011) and Schaefer et al. (2009)
Full size table

Local 87Sr/86Sr Isotopic Footprint

The 87Sr/86Sr from both new fauna enamel samples and available human femurs (Fig. 4, Table 3; for complete database and references see Online Resource SI 1: Table 1 and 2; SI 2: Fig. 1) indicate that the local isotopic range at Těšetice-Kyjovice and Vedrovice presents similar strontium values (0.71042–0.71249) (excluding an Ovis/Aries outlier from Těšetice-Kyjovice with higher radiogenic values = 0.71651, which could be the result of transhumance or of being a trade item), as they both lie on the border between the Carpathian basin and the Bohemian Massif. The values from Brno-Starý Lískovec, situated on the Bohunice plateau, are slightly lower (0.70973–0.71014) (see complete database and references in SI 1: Table 1 & 2), probably because the loess in this area derives from late Weichselian interpleniglacial soil complex with varied underlying geology defined by a wide range of granitoids (Starkel, 1977; Mateiciucová, 2008). Accordingly, while the Moravian groups settled on loess soils, as was common among the LBK communities (Bentley & Knipper, 2005a; Bentley et al., 2006; Price et al., 2002, 2004), the higher 87Sr/86Sr ranges obtained in this area in comparison with the regular loess 87Sr/86Sr ratios (0.70860–0.71030) suggests that there is a greater contribution from Precambrian and Proterozoic rocks to the loess in the vicinity of the sites, which may reflect their position on the edge of the Bohemian–Moravian Highlands, where the strontium values are higher.

Fig. 4
figure 4

Local isotopic signature based on available human femurs’ 87Sr/86Sr data and new faunal samples. Abbreviations of the sites are as follows: Těšetice = Těšetice-Kyjovice; Brno = Brno-Starý Lískovec; Za Dvorem = Vedrovice-Za Dvorem; Siroká = Vedrovice-Široká u Lesa; Sídliště = Vedrovice-Sídliště. The Sus sample from Vedrovice was published by Smrčka (2019) without specifying to which sector it belonged. Graph produced by Ggplot2 software

Full size image
Table 3 Regional 87Sr/86Sr isotopic signatures of the analysed sites
Full size table

87Sr/86Sr from south-eastern Moravian human, geological and animal samples (see Online Resource SI 1: Table 3) were combined to produce strontium isotopic baselines for each of the geological formations in the region. The observed ranges are as follows (Table 4): for Tertiary and Quaternary soils from the Carpathian Foredeep (TQCF), the isotopic range is approximately 0.71054–0.71056 when considering geological samples alone. However, when incorporating measurements from the primary human cluster, the range expands to 0.71018–0.71080. The Paleozoic formations in the Bohemian Massif (Paleo BM) exhibit a range of 0.70946–0.71059. In the Quaternary soils behind Ždánický les mountains (Quat BPZ), the range is 0.70941–0.71038, and for Ždánický les itself (Tertiary Z) it is 0.70891–0.70979. The Proterozoic from the Bohemian Massif (Prot BM) baseline was built on the basis of the local isotopic range from Těšetice-Kyjovice and Vedrovice, measured on human femurs and domestic fauna (0.71042–0.71249) and matching the data with the human long bones and Sus scrofa samples from the sites of Krumlovský les and Trstěnice (Fig. 5).

Table 4 87Sr/86Sr local isotopic footprint from south-eastern Moravia
Full size table
Fig. 5
figure 5

Strontium isotopic values (87Sr/86Sr) from Moravian human, geological and animal samples, including those from the studied archaeological sites. Paleo BM = Paleozoic from the Bohemian Massif; Prot BM = Proterozoic from the Bohemian Massif: TQCF = Tertiary and Quaternary soils from the Carpathian Foredeep; Quat BPZ = Quaternary soils behind Ždánický les mountains; Z = Ždánický les mountains (Carpathians); m = males; f = females. See SI-2: Fig. 1 to check the animal species, and SI-2: Fig. 3 to check the isotopic values for the humans. Graph produced by Ggplot2 software

Full size image

87Sr/86Sr, Sex and Age

Considering the 87Sr/86Sr values, in all the sampled sites except Brno-Starý Lískovec, the non-local values are predominantly found in female individuals and the local ones in males and non-adults (Fig. 6). Most of the Vedrovice-Široká u Lesa, Vedrovice-Za Dvorem, Vedrovice-Sídliště and Těšetice-Kyjovice local individuals’ 87Sr/86Sr values fall within the isotopic footprint of the Proterozoic Bohemian Massif, where the sites are located.

Fig. 6
figure 6

Dot plot including the different sites’ 87Sr/86Sr distribution according to males’ and females’ molars M1, M2 and M3 (excluding the male outlier from grave 69/78 from Vedrovice-Siroká u Lesa). The grey shading indicates the local isotopic footprint. Graph produced by Ggplot2 software

Full size image

At Vedrovice-Sídliste, where only males and non-adults were recorded, all strontium values matched the local baseline. At Vedrovice-Za Dvorem, Vedrovice-Široká u Lesa and Těšetice-Kyjovice, females displayed a greater isotopic variability than males. Only 2 male individuals were classified as non-local in the whole sample, both buried at Vedrovice-Siroká u Lesa (11% of the total sampled males). This contrasts with 2 non-local females at Těšetice-Kyjovice (66% of the sampled females), 3 at Vedrovice-Za Dvorem (50% of the sampled females) and 13 at Vedrovice-Siroká u Lesa (41% of the sampled females) (Fig. 6). Thus, a strong statistical relationship between biological sex and mobility is observed from the aggregated data (see χ2 test results in Online Resource SI 2: Table 2).

In the case of Brno-Starý Lískovec, where only males and non-adults were buried, two males presented local values (burials 805 and 806), whereas three displayed ranges outside the local (burials 800, 801 and 803). In this case, the sampled molars corresponded to M1 and M2, indicating that these individuals probably moved during infancy.

The different 87Sr/86Sr ratios from the three molars (M1, M2 and M3) were compared by sex in the cases of Vedrovice-Široká u Lesa, Vedrovice-Za Dvorem, Vedrovice-Sídliště and Těšetice-Kyjovice. As a result, only in the case of M3 (mineralized between 7 and 17 years) was the relationship between sex and mobility statistically significant (see Online Resource SI 2: Table 2). In the cases of M1 (mineralized during the last month of gestation up to 2 years) and M2 (mineralised between 1 and 8 years), no statistically significant sex-related patterns were observed (see Online Resource SI 2: Table 2), even though there were more non-local females (4 M1, 8 M2, 8 M3) than males (0 M1, 1 M2 and 2 M3). This means that mobility patterns for males and females were especially divergent at the point the M3 was forming (7–17 years), potentially linking greater female mobility to the beginnings of puberty and early adolescence.

A closer analysis of the non-local 87Sr/86Sr values indicates the presence of potentially different mobility groups. Most of the non-local individual signatures do not differ much from the isotopic footprint of the other geological areas nearby: the Vedrovice (including the three sectors) and Těšetice-Kyjovice non-local females present 87Sr/86Sr values between 0.71228 and 0.71048, which match the Bohemian Massif Palaeozoic baseline as well as the Quaternary and Tertiary sediments from the Carpathian Foredeep, the Ždánický les mountains (corresponding to the beginning of the Carpathians), and, especially, the Quaternary soils east of the Ždánický les. Even though the 87Sr/86Sr analysis interpretative limits do not allow us to assess where exactly these females were moving, the similarities with the nearby geological deposits of the region suggest that they could be involved in short-distance movements within this area (nearby areas located less than 50 km away). Yet, a male and female individual from Vedrovice-Siroká u Lesa (graves 99/81 and 29/76) are exceptions with extreme non-local values that do not match any of the baselines in south-eastern Moravia and must have arisen in movement from more distant regions.

In the case of the three non-local males from Brno-Starý Lískovec, isotopic values match the footprint defined for the Vedrovice complex, being clearly beyond the Bohemian Massif palaeozoic baseline and within the Bohemian Massif palaeozoic range (Figs. 5 and 6), and suggesting that this could have been the area through which they were moving.

The 87Sr/86Sr values of Vedrovice and Těšetice-Kyjovice fall within the Proterozoic isotopic footprint and within the Paleozoic at Brno-Starý Lískovec; thus, the so-called ‘locals’ could either have moved westwards within the Bohemian Massif or not moved at all from their settlements located at the border between the Carpathian basin and the Bohemian Massif.

δ 18O and 87Sr/86Sr

The correlation between δ18O and 87Sr/86Sr isotopic data contributed interesting results (Fig. 7), compensating for the interpretative limitations of 87Sr/86Sr analysis. In the case of samples which have local 87Sr/86Sr determinations and δ18O data (see Online Resource SI 1: Table 1), the isotope local/non-local attributions agree. Brno-Starý Lískovec samples are considered local though 87Sr/86Sr analysis presented slightly different δ18O values (− 7.15/-7.45) from Vedrovice’s (including the three sectors) and Těšetice-Kyjovice’s (− 6.3/− 7.45), probably because of the consumption of water from different Svratka river tributaries (see Online Resource SI 2: Fig. 3).

Fig. 7
figure 7

Biplot indicating the relationship between the δ18O and 87Sr/86Sr data from the analysed sites according to the sites A and the sampled individuals’ sex and animal species B (excluding the Ovis/Aries outlier from Těšetice-Kyjovice with higher radiogenic values). Graph produced by Ggplot2 software

Full size image

In the case of the individuals for whom we have non-local 87Sr/86Sr determinations and δ18O data (4 in total), grave 11 from Těšetice-Kyjovice, and 10/89 and 7/88 from Vedrovice-Za Dvorem presented δ18O values similar to those belonging to Brno-Starý Lískovec, which suggests this area as their probable place of origin. At the same time, the individual from Brno-Starý Lískovec considered as non-local on the basis of 87Sr/86Sr (burial 803), also presents δ18O ranges similar to those determined at Vedrovice and Těšetice-Kyjovice (Fig. 7b). This probably indicates short-distance movements between the two areas, which are geologically different (and hence have distinct 87Sr/86Sr footprint), but with similar water resources (and similar δ18O values).

In the case of the fauna samples used in 87Sr/86Sr analysis to identify the local signature, they present clearly different δ18O values from those which can be considered local in all sites (Fig. 7a: see Online Resource SI 1: Table 1). At Těšetice-Kyjovice, Sus and a Bos have relatively high values (between -8 and -9), at Vedrovice-Za Dvorem, Ovis presented very low values (-5.3) and Bos high ones (-8) and, finally, at Brno-Starý Lískovec, Bos presented very low values (-5.8) and Ovis high ones (-8). This last data has been interpreted bearing in mind that δ18O composition of tooth enamel is species dependent and so the measured values of the various species cannot be compared directly.

Mobility, Diet and Grave Goods

Multiple ribs and long bone samples have been analysed for dietary isotopes (δ15N and δ13C) in the framework of other projects, confirming an essentially terrestrial diet based on C3 cereal in all sites (Whittle et al., 2013). Bearing that in mind, the available δ15N data variation (see raw data at Online Resource SI 1: Table 2) was used to chart diversity in protein consumption. The δ15N values were statistically different between male and female individuals considering all sites together (T Test 0.0005462, F Test 0.23844 for normal distributions after checking with Shapiro Wilk: see details at Online Resource SI 2: Table 3), with females’ values clearly lower than those of males (Fig. 8). These results could be an indicator of different rates of protein consumption, with males eating more protein than females, or due to the consumption of animals from other areas with different manuring systems or no field manuring at all (as suggested by Richards et al. (2008) and Masclans et al. (2021a, 2021b) for Vedrovice-Široká u Lesa).

Fig. 8
figure 8

Dot plot including δ15N values according to sex and site. Abbreviations: f = females, m = males. Graph produced by Ggplot2 software

Full size image

Two δ15N clusters were identified using K-means (see Online Resource SI 2: Fig. 4), including ‘high’ (9.8–10.8, including 19 males, 11 females) and ‘low’ (8.5–9.7, including 8 males, 29 females) δ15N values, and used in the MCA analysis.

According to χ2 test, there is not a significant variation in the presence of furnished or unfurnished graves between males and females (χ2 test = 0.006835; df 1; p no assoc = 0.93411), nor between the different funerary sites (χ2 test = 3.341; df 4; p no assoc = 0.50246, in this case including also the non-adult individuals).

To examine whether sex and mobility (87Sr/86Sr & δ18O) of the buried individuals co-varied with the grave goods distribution and the available isotopic dietary values (δ15N), a correlation between these factors was explored through Multiple Correspondence Analysis (MCA). The MCA results indicate a relationship between the sex of the buried individuals, the grave goods assemblages, diet and mobility (see Online Resource SI 2, Figs. 4 & 5 for full statistics, and SI 4 for the database) (Fig. 9). Males with higher δ15N dietary values were more likely to be buried with more stone and bone tools, as well as pottery vessels, whereas males with lower δ15N dietary determinations were correlated with a lesser presence of grave goods.

Fig. 9
figure 9

Multiple Correspondence Analysis of the buried individuals’ sex, mobility (87Sr/86Sr), diet (δ15N) and grave good distribution. Graph produced by Ggplot2 software

Full size image

In contrast, female individuals do not present statistical associations with any particular kind of grave good. In this case, non-local females were less likely to receive grave goods (26% local unfurnished, 41% non-local unfurnished), although no differences were identified in terms of the variability of the objects. Non-local females presented generally less δ15N variability than local ones (values between 9 and 10), with statistically different value distributions (Kolmogorov–Smirnov 0.0064).

Discussion and Conclusions

The results of this study confirm that sex and age played a key role in the mobility patterns of the LBK Moravian communities: not only were females more mobile than males, but their mobility increased progressively with age. The M3 are statistically more likely to show a non-local strontium isotope among females than males. As the mineralisation of the third molar occurs at roughly the same time as puberty, we can suggest that female mobility significantly increased from puberty and early adolescence.

We also observed that the mobility patterns do not differ between the various funerary contexts (settlements and cemeteries), suggesting that what determines the presence of non-local individuals is not the funerary pattern but the proportion of females versus males in each of the sites.

At least four possible mobility realities coexisted at the studied sites. The first one corresponds to ‘local’ males, females and non-adults. Considering that the ‘local’ 87Sr/86Sr baseline falls within the Proterozoic isotopic footprint at Vedrovice and Těšetice-Kyjovice and within the Paleozoic at Brno-Starý Lískovec, they could either have been moving westwards within the Bohemian Massif or not moving at all from their settlements located at the border between the Carpathian basin and the Bohemian Massif. However, the hypothesis that they did not move at all from their place of settlement becomes less plausible in the light of δ18O data, which supports the idea of short-distance movements by ‘local’ individuals consuming water from the same Svratka river tributary.

In contrast, the non-local females were possibly moving eastward at distances of less than 100 km (from the Carpathian Foredeep, the Carpathian Ždánický les mountains or the Quaternary soils east of the Ždánický les) and/or from the north (Paleozoic of the Bohemian Massif). Despite the similarity of most of the non-local values with the nearby areas, the hypothesis that they moved to eastward regions with similar 86Sr/87Sr isotopic determinations cannot be ruled out.

This could also have been the case with some non-local females from Vedrovice-Široká u Lesa and Vedrovice-Za Dvorem, whose values coincide with the local isotopic footprint of the contemporary Slovak cemetery of Nitra (between 0.70932 and 0.70964) (Masclans et al., 2021b), with an isotopic footprint close to the Carpathian Mountains and at a distance of about 180 km. However, the δ18O isotopic data, though still very sparse, points towards the first hypothesis (that is, eastward movements at distances under 100 km), as both individual 11 from Těšetice-Kyjovice and 10/89 from Vedrovice-Za Dvorem, considered as non-local based on Strontium values, present δ18O values similar to those belonging to Brno-Starý Lískovec, located at the border of the Paleozoic of the Bohemian Massif, indicating short-distance movements between geological areas with different Strontium isotopic baselines but similar water resources.

Finally, only in two cases are there extreme non-local values that do not match any of the baselines in south-eastern Moravia: one female and one male, from Vedrovice-Siroká u Lesa. This kind of mobility thus seems to be not sex-related and is clearly rare.

The anthropological record shows an unequal distribution of males/females according to the funerary record (female-sexed skeletons outnumber males at cemeteries and burial clusters, and the situation is reversed at the settlement graves), indicating selection based on sex in burial practices. This selection is seen in the age of buried individuals, especially among males. The unbalanced presence of buried men in the funerary contexts could be explained by the presence of differential access to certain funerary practices defined by sex and age, but it could also indicate some kind of age-based mobility pattern among men.

Traditionally, female migration, patrilocality and community exogamy have been considered as a factor explaining the higher proportion of females than males in burial contexts, as well as a more plausible explanation for higher variability of 87Sr/86Sr ratio in females than males (Chamberlain, 2006; Hrnčíř et al., 2020a; Whittle et al., 2013). Our results, however, challenge this idea and lead to new hypotheses no longer focused on whether or not females moved for reasons of exogamy, but rather why there is a more mobile group of females and what differentiates them from the rest of their community.

These results challenge the dichotomy between ‘locals’ and ‘non-locals’, which was previously questioned by Bickle and Hofmann (2007) and Snoeck et al. (2016); for further discussion about the problems related to what can be considered ‘local’ or ‘non-local’ in prehistory see Cavazzuti et al. (2019). Instead of ‘local’ versus ‘non-local’, we suggest here that it may be more accurate to speak about mobile versus non-mobile individuals. It makes sense to interpret those non-local females as simply more mobile than the rest of their community, bearing in mind that this mobility was short-distance in nature.

As the strontium isotope record of LBK funerary contexts has grown, it has demonstrated a heterogeneous patchwork of mobility patterns, raising questions as to the existence of a single predominant LBK patrilocal model. In south-western Germany, for example, isotopic data suggested differences between male and female individuals, pointing to major mobility among males, which has been interpreted as the result of the integration of marginal landscapes (uplands) for animal husbandry (Bentley & Knipper, 2005b; Knipper, 2009). In settlements like Stuttgart-Mühlhausen (Baden-Württemberg, Germany) different mobility patterns were identified among the cemetery sectors, as well as between the sexes according to their ages (Knipper, 2011). A different situation was also found at the settlement of Nieder-Mörlen (Hessen, Germany), where the individuals who died as juveniles presented non-local values, while those who died as adults had local ones (Montgomery et al., 2009).

This kind of data re-interpretation agrees with the latest aDNA studies on two LBK cemeteries, Nitra and Derenburg-Meerenstieg II (Childebayeva et al., 2022; Gelabert et al., 2023), where only about 60% (in Nitra) and 25% (in Derenburg-Meerestieg II) of the individuals had biological relatives buried at the same site. This low degree of kinship relationships at cemeteries suggests that it was not only the individuals determined as non-local by 87Sr/86Sr who moved from their places of origin, but also the local ones, both female and male, since they were not genetically related either. Together this data makes us rethink who is being buried in these contexts and what defines them as a group that belonged together in death. It opens the possibility of understanding burial groups no longer as a straightforward representation of a living community and its kinship structure, as has been implicitly assumed for previous models of patrilocality.

In order to create more data to better understand those questions, we explored the correlation between the δ15N dietary values and the number of artefacts deposited as grave goods. As a result, we observed that the more mobile females were less likely to be buried with grave goods (or, at least, with non-perishable grave goods) and statistically more related to lower δ15N values than non-mobile females, who, in turn, presented lower δ15N values and lower grave-good variability than male individuals.

The lower δ15N values associated with these more mobile females could either reflect different dietary practices involving lower animal protein ingestion rates, or the consumption of animals from other areas with different manuring systems or no field manuring at all. One possible explanation for this phenomenon may be related to economic practices involving seasonal movement between geological areas (as indicated by 87Sr/86S analysis), but at short distances with the same climatic conditions and very similar water resources (as suggested by oxygen analysis). The presence of transhumance activities within regions presenting similar strontium isotopic baselines could be an explanation for this variability.

Especially striking is the fact that the different samples of Ovis and Bos from the same sites present different δ18O values. This variability could be related to the consumption of different hydric resources or because of differences in the animals’ diet (Pederzani & Britton, 2019). The presence of transhumant activities within regions presenting similar 87Sr/86S isotopic baselines could be an explanation for this δ18O variability.

There may also be a division of labour by age and sex associated with mobility. The sexual division of labour has already been proposed in relation to the distribution of tools deposited as grave goods within cemeteries (Masclans et al., 2020, 2021a, 2021b), while osteological and physical examination of upper and lower limbs reveals musculoskeletal stress markers which strongly suggest sex-differentiated activity patterns and labour-related asymmetry between the sexes (Berner et al., 2018; Macintosh et al., 2014a, 2014b, 2017; Masclans et al., 2021b). We find it relevant that this more mobile form of life is directly associated with the onset of puberty, contradicting accepted models so far, which mostly related mobility to M2 (Bickle & Whittle, 2013; Masclans et al., 2020). That is, it would appear that, for females, the onset of a more mobile lifestyle corresponded with the transition to adulthood and the assumption of responsibilities within the community, and which involved a marked sexual division of labour. However, more evidence needs to be collected to confirm this hypothesis, especially data concerning animal husbandry and mobility practices related to transhumance, as well as more detailed research on osteology and musculoskeletal stress markers.

Although our results do not directly explain the mobility of this group of females, they do shed light on the wide range of possible reasons for it, as well as on the complexity of the social organisation of these communities. Our study shows that the female–male binary and patrilocal exogamic practices are not on their own sufficient as an analytical framework to interpret the mobility patterns of these central European first farmers.

Overall, we demonstrate that a sedentary lifestyle did not necessarily imply that the groups remained static; rather, they changed their mobility strategies as a result of population migrations, which expanded the new economic model in most of the European regions to smaller-scale mobility potentially related to kinship or labour.