Diverse ancestries

The early genome-wide studies set the fundamental contours of Neolithic development by arguing for two major migration events in central Europe (Allentoft et al. Reference Allentoft and Willerslev2015; Haak et al. Reference Haak and Reich2015; recently Allentoft et al. Reference Allentoft and Willerslev2024b). However, the excitement surrounding the now available scientific toolkit should not seduce us into an illusion that we see more clearly into the past (Kristiansen Reference Kristiansen2014). Genetic methods ultimately depend on archaeology as a provider of samples and are therefore vulnerable to the same limitations. We need to admit that archaeological inquiry is unavoidably shaped by survivor bias – a cognitive delusion that prompts researchers to ponder only what passed the selection of archaeological transformations, while ignoring many aspects of past lives which did not (Lucas Reference Lucas, Chapman and Wylie2015a). Indeed, one can hardly work with missing data, but archaeologists in particular need to bear in mind that the excavated record is, by its nature, largely incomplete.

Genetic studies depend on the preservation of funerary contexts, preferably those containing inhumation burials. Yet such contexts are underrepresented or entirely absent for many periods and regions of the Neolithic in central Europe. Even where inhumations are available, sampling may be subject to additional biases. As Anna Källén (Reference Källén2025, 48) notes, several factors render aDNA sampling far from a randomised analytical process. Particularly in the early stages of methodological development, when these analyses were a high-risk endeavour which required substantial funding, applications of genomic science were guided by pragmatic considerations aimed at mitigating potential failures. Researchers tended to sample well-contextualised, artefact-rich, and well-preserved archaeological relics, which promised reliability and complementary contextual information. Bioarchaeological examinations of past individuals, including aDNA analyses, are still most commonly applied to larger cemeteries, where the number of burials provides a sufficient sample base and abundant grave goods allow for additional labels on these samples, specifying chronology, cultural affiliations, social status, and other aspects. These cemeteries are generally assumed to be simply a window offering insight into a discrete settlement community, essentially a village or a cluster of neighbouring villages. Yet such an assumption rests on an uncritical preconception shaped by over a millennium of Christian burial tradition, in which the parish church cemetery functions as the dominant model. Alternatives need to be considered (e.g. the regional model by Gaydarska et al. Reference Gaydarska, Roe and Slavchev2025).

Linearbandkeramik cemeteries, for instance, have been recently revealed as a rather short-lived phenomenon (Morell-Rovira et al. Reference Morell-Rovira, Bickle, Hamilton, Díaz-Zorita Bonilla, Francken and Masclans2025). Genome-wide analysis of one such cemetery in Nitra, Slovakia, yielded only highly fragmentary pedigrees (Gelabert et al. Reference Gelabert and Reich2024), with first- and second-degree relations appearing less frequently than one would expect for a patrilocal settlement community. Incomplete kinship structures stand in sharp contrast to the more coherent pedigrees identified for later Neolithic burial grounds (Krause-Kyora et al. forthcoming; Rivollat et al. Reference Rivollat, Haak, Meller, Krause, Haak and Risch2023; Seersholm et al. Reference Seersholm and Sikora2024). Beyond the possibility that family members were dispersed across different burial sites due to higher mobility, this pattern can also mean that Linearbandkeramik cemeteries offer a highly selective assemblage of individuals. Shaped by unknown criteria or aggregating deceased from large distances (J. Chapman et al. Reference Chapman, Gaydarska and Jakob2024), they may thus underrepresent certain social segments. It is very likely that a considerable part of the population was buried outside these cemeteries (Pechtl & Hofmann Reference Pechtl, Hofmann and Müller-Scheeßel2013).

Despite the sampling biases inherent to aDNA research, there is reason for optimism, as the field has entered a self-correcting phase. While early hints of genetic diversity (see e.g. Haak et al. Reference Haak and Reich2015; Hofmanová et al. Reference Hofmanová and Burger2016) were downplayed in favour of grand narratives stressing sweeping migratory waves, rapid disciplinary expansion has produced larger datasets enriched by locally focused, finer-grained studies, which offer more complex insight into the genetic composition of past populations. Particularly striking is the persistence and resurgence of European Mesolithic ancestry (here I deliberately use geographic-temporal designations and avoid labels such as Western hunter-gatherers or Neolithic farmers, which straightforwardly conflate genetic ancestry with subsistence strategies; see Eisenmann et al. Reference Eisenmann and Stockhammer2018).

A population with a Mesolithic-derived genetic signature is now detectable across Europe for centuries after the purported large-scale migration of groups bearing Anatolian-derived ancestry (Allentoft et al. Reference Allentoft and Willerslev2024a; Brunel et al. Reference Brunel and Pruvost2020; Carvalho et al. Reference Carvalho, Fernández-Domínguez, Arroyo-Pardo, Robinson, Cardoso, Zilhão and Gomes2023; Rowley-Conwy Reference Rowley-Conwy2024; Simões et al. Reference Simões and Jakobsson2024). Specifically in central Europe, this trend was demonstrated by Mark Lipson and colleagues (Reference Lipson and Reich2017), who identified a substantial admixture of Mesolithic ancestry in Middle Neolithic populations from Hungary and Germany. At a more local scale, the Blätterhöhle cave (Figure 1) provides one of the most compelling cases, owing to the combined application of genetic and stable isotope analyses (Bollongino et al. Reference Bollongino, Nehlich, Richards, Orschiedt, Thomas, Sell, Fajkošová, Powell and Burger2013; Orschiedt et al. Reference Orschiedt, Bollongino, Nehlich, Burger, Schier, Orschiedt, Stäuble and Liebermann2021). Three individuals interred within the cave exhibit diets based on terrestrial products while retaining 40–50% Mesolithic ancestry. Another individual, who consumed a hunter-gatherer-fisher diet, remained genetically closer to the European Mesolithic population, but also carried 27% Anatolian-derived ancestry. Admixture modelling indicates relatively late gene flow between populations displaying Mesolithic and Anatolian DNA signatures, occurring around 4000 BCE. Additional evidence from the German lower mountain ranges comes from communities associated with the Wartberg culture (3500–2800 BCE). A total of 147 individuals buried in five gallery graves (Altendorf, Niedertiefenbach, Rimbeck, Warburg, and Züschen) carried 22–56% Mesolithic ancestry, likely introduced into the population between 4500 and 3500 BCE (Da Silva et al. Reference Da Silva and Krause-Kyora2025; Immel et al. Reference Immel and Krause-Kyora2021; Krause-Kyora et al. forthcoming). A comparably late admixture process has been documented in Bohemia, where the Neolithic burial sites sampled by Luka Papac and colleagues (Reference Papac and Haak2021) revealed that Mesolithic ancestry increased up to 20% on average, with introgression clearly under way during the 5th millennium BCE.

Figure 1.

Map showing the regions and sites discussed in the text. Dashed lines mark the approximate extent of the Linearbandkeramik culture, associated with the spread of farming population in the late 6th millennium, and the Corded Ware and Yamnaya cultures, linked to early steppe-origin migration in the first half of the 3rd millennium. 1 Blätterhöhle, 2 Altendorf, 3 Niedertiefenbach, 4 Rimbeck, 5 Warburg, 6 Züschen, 7 Stupno-Břasy, 8 Voitmannsdorf-Stroholz, 9 Kirschbaumhöhle, 10 Wattendorf-Motzenstein, 11 Burgerroth, 12 Švarcenberk, 13 Pobiel 10, 14 Glanów 3, 15 Łykowe 1, 16 Mokracz 1.

Figure 1 Long description

The map of central Europe highlights regions and sites important to this study. The dashed orange line indicates the approximate extent of the Linearbandkeramik culture, associated with the spread of farming populations in the second half of the 6th millennium. The dashed green line shows the extent the Corded Ware and Yamnaya cultures, linked to early steppe-origin migration in the first half of the 3rd millennium. Key regions of diverse trajectories such as Muskau Heath, Franconian Alb, West Bohemia, the Upper Warta Basin, and others are marked with red shaded areas. Sites discussed in the text, like Blaetterhoehle, Altendorf, Niedertiefenbach, Rimbeck, Warburg, and others are labeled with numbers. The map provides a visual representation of the patterns of diversity in Neolithic central Europe.

Detailed aDNA data also challenge the narrative that Late Neolithic populations carrying Anatolian ancestry were completely replaced by migrants of Pontic steppe origin in the early 3rd millennium. In Bohemia, the earliest individuals associated with steppe-related ancestry intermarried with women whose genetic profiles do not align with the local Late Neolithic population. Nonetheless, pre-steppe Neolithic ancestry reappeared much later: first around 2400 BCE in Bell Beaker graves, and again in the early 2nd millennium BCE as part of the genetic makeup of populations documented at the onset of the Bronze Age (Papac et al. Reference Papac and Haak2021). A comparable pattern is observed among Late Neolithic to Middle Bronze Age inhabitants of the Lech Valley in Bavaria (Mittnik et al. Reference Mittnik and Krause2019). Here, the arrival of steppe ancestry does not resemble a sudden demographic break, but rather a pulse-like admixture event whose overall impact diminished over the long term.

Although advances in aDNA sampling reveal substantial persistence or reappearance of earlier ancestries long after they were thought to have been erased by subsequent migrations, these results rarely indicate where the source populations may have come from. Some authors argue that such ancestries were reintroduced through demic expansions from the outskirts of areas affected by major migrations, where archaic ancestries remained available within local groups (Beau et al. Reference Beau, Rivollat, Réveillas, Pemonge, Mendisco, Thomas, Lefranc and Deguilloux2017; Da Silva et al. Reference Da Silva and Krause-Kyora2025; Immel et al. Reference Immel and Krause-Kyora2021; Papac et al. Reference Papac and Haak2021). While this scenario may hold in certain cases tested through admixture analyses (e.g. multiple sources of Mesolithic ancestry identified in various Neolithic populations by Papac et al. Reference Papac and Haak2021, cf. Brami Reference Brami, Heitz, Wunderlich, Hinz and Furholt2023, who highlights the limitations of these tests), it remains largely deductive in most other instances. More importantly, it implicitly reproduces the assumption that ‘otherness’ can be situated only at the outer frontiers, rather than within the interior landscapes of central Europe. Yet, as the following sections demonstrate, other social worlds exhibiting considerable diversity were situated here, rendering such generalist assumptions unnecessarily restrictive.

Diverse material representations

Central European archaeologists have traditionally emphasised pottery – primarily its decoration and morphology, rather than manufacturing techniques – as a material representation of identity and belonging. This predisposition persists despite ethnographic research demonstrating that a much broader spectrum of objects can encode such meanings. Deliberate or habitual choices for domestic architecture and its patterns, weapons, or personal ornaments, to name just a few archaeologically visible categories, are equally able to signal various group affiliations (e.g. Hodder Reference Hodder1982; Wiessner Reference Wiessner1983). Nevertheless, the scope of archaeological inquiry is further narrowed by pre-understandings about how material representations ought to be structured in time and space (Wylie Reference Wylie2017). Although the idea of rigid spatio-temporal sequencing of monothetic archaeological cultures has been under fire for a long time (e.g. Shennan Reference Shennan and Shennan1989, with literature), it continues to function as an implicit yet powerful framework for assessing prehistory in central Europe (Šabatová Reference Šabatová2013). Patterns that fall ‘outside the box’, such as highly mixed distributions of characteristic objects, are typically sought on frontiers and contact zones of cultures, but far less readily anticipated within the putative heartlands of distinctive material traditions. This framework-induced blindness can be amplified when archaeologists are confronted with quantitatively and qualitatively sparse evidence in certain regions.

The regions of south and west Bohemia yield little evidence of occupation throughout most of the 4th millennium BCE, but sites increase markedly towards its end (John Reference John, Vondrovský and Chvojka2021; Metlička Reference Metlička2015). Assigned to the Řivnáč and Cham cultures, they were part of a broader central European Late Neolithic horizon characterised by shared ceramic attributes, such as surface roughening or plastic cordons, and settlements of semi-sunken houses, which were often fortified and situated at elevated locations, mostly rocky hilltops and river promontories (e.g. Bogner Reference Bogner2017; Engelhardt Reference Engelhardt and Schmotz2002; Gohlisch & Reisch Reference Gohlisch and Reisch2001; Medunová Reference Medunová1977; Zápotocký & Zápotocká Reference Zápotocký and Zápotocká2008; Zuber Reference Zuber, Chvojka, Chytráček, Gruber, Husty, Michálek, Sandner, Schmotz and Traxler2019). Nearly 80 such sites are known from west Bohemia and about ten from south Bohemia (John Reference John2010; Reference John, Vondrovský and Chvojka2021). In line with the general framework of the Řivnáč and Cham cultures in Bohemia, they were initially dated to 3200–2800/2700 BCE (Zápotocký Reference Zápotocký and Neustupný2013). Only thanks to radiocarbon dating, often targeting macroremains of cultivated plants, has it been revealed that communities in south and west Bohemia continued to inhabit these sites and maintain this pottery tradition until at least 2400 BCE. This is about three centuries longer than neighbouring regions of central Bohemia, eastern Bavaria, and Moravia (Vondrovský et al. forthcoming).

The persistence of the hilltop-settlement traditions was accompanied by minimal integration of a new set of material elements, most prominently shaft-hole battle axes and cord-impressed beakers, emerging primarily in single-grave cemeteries across central Europe in the early 3rd millennium BCE. Instead of subsuming them under a cohesive entity such as a Corded Ware ‘folk’, this material manifestation has been emphasised as an idealised set permeating different communities and adopted in a variable manner (Furholt Reference Furholt2014b; Reference Furholt2019; Wentink Reference Wentink2020). It should not be surprising that some conformed reluctantly or largely refused. South and west Bohemia provide no substantial evidence of this material repertoire and single-grave burials, even though such cemeteries appear in north-western and central areas of Bohemia as early as 2900 BCE (Dobeš et al. Reference Dobeš, Pecinovská, Ernée, Heyd, Kulcsár and Preda-Bălănică2021). From a culture-historical viewpoint, south and west Bohemia were thus regarded as largely uninhabited during the ‘Corded Ware period’ (Menšík Reference Menšík, Chvojka, Chytráček, Gruber, Husty, Michálek, Sandner, Schmotz and Traxler2017). The Corded Ware as well as later Bell Beaker material representations in these regions involve only finds of a few uncontextualised pottery fragments and stray battle axes (John Reference John, Vondrovský and Chvojka2021; Metlička et al. Reference Metlička, Řezáč and Turek2007). Their interpretive potential is limited; in particular, the latter may have been manipulated or otherwise redistributed as so-called thunderstones during later, even sub-modern periods. Nevertheless, at least one case is indicative. A typical battle axe head recovered from the hilltop site of Stupno-Břasy suggests that west Bohemian hilltop communities were likely aware of the new material representations, exhibited at cemeteries in neighbouring regions.

The late 4th to early 3rd millennium BCE saw settlement and land use intensification also in the northern Franconian Alb, Baravia (Fuchs et al. Reference Fuchs, Will, Kunert, Kreutzer, Fischer and Reverman2011; Kothieringer et al. Reference Kothieringer, Seregély, Jansen, Steup, Schäfer, Lambers and Fuchs2023; Pechtl Reference Pechtl and Metlička2024). Late Neolithic material elements (Bernburg – Burgerroth – Cham – Goldberg III – Wartberg) persisted here well into the 3rd millennium (overview in Nadler Reference Nadler, Uthmeier and Mischka2023, 912–25). For example, excavations at Voitmannsdorf-Stroholz revealed a settlement at an elevated location with a characteristic pottery and lithic assemblage that lacks clear Corded Ware elements, despite a radiocarbon date on animal bone indicating occupation at least until 2600–2300 BCE (Dürr et al. Reference Dürr, Müller, Riedmüller, Schulz, Seregély and Tillmann2004). At that time, Corded Ware representations were well established (particularly through characteristic burials) in neighbouring lowlands (Seregély Reference Seregély2008; Ullrich Reference Ullrich2008; Vondrovský et al. forthcoming), and early Bell Beaker sets emerged in north-west Bavaria around 2550 BCE (Ullrich Reference Ullrich2008). During this time, communities in the northern Franconian Alb also continued the local tradition of cave burials, which can be traced back to the Mesolithic and Early Neolithic (Richter Reference Richter, Uthmeier and Mischka2023, 669–72; Seregély Reference Seregély and Falkenstein2012). In the Kirschbaumhöhle cave, deposition of human bodies took place in 2900–2600 BCE (Seregély et al. Reference Seregély, Burgdorf, Gresik, Müller and Wilk2015).

Yet, other sites testify to a certain integration of Corded Ware objects and material representations. These have been occasionally recorded in the Franconian Alb in rather non-conventional contexts, such as rock shelters and striking rock formations (Falkenstein Reference Falkenstein2012). An intriguing mixture of cultural elements can be seen at Wattendorf-Motzenstein, occupied between 2630 and 2470 BCE. Although presented as an example of a Corded Ware culture settlement (Müller & Seregély Reference Müller and Seregély2008; Seregély Reference Seregély2008), the site retains key Late Neolithic attributes: a hilltop location with sunken-floor houses. Corded Ware elements were nonetheless manifested in pottery, both in its style and manufacturing technology. In this respect, the Franconian Alb parallels some settings along the western frontier of the Corded Ware distribution in the Alpine foreland, where cord-impressed pottery was incorporated into the traditions of local lake dwellers (Suter Reference Suter2017). It is worth pointing out, however, that chronological overlap of Late Neolithic and Corded Ware material representations is generally accepted for southern Germany (Dunne et al. Reference Dunne, Szilágyi, Casanova, Griffiths, Knowles, Evershed and Hofmann2023; Furholt Reference Furholt2003), and a comparable mixture is well-documented 80 km south-west of the Franconian Alb at Burgerroth, a fortified hilltop settlement inhabited until the 26th century (most recently Link Reference Link, Hofmann, Mischka and Scharl2025). The Franconian Alb is thus best understood as a spatially distinctive example of processes that took place in a broader area of the German lower mountain ranges during the 3rd millennium (Drummer Reference Drummer, Hofmann, Mischka and Scharl2025).

On a more general note, these cases reveal communities that remained committed to their established material representations, specifically pottery production, settlement location, and architecture, even though they were exposed to new influences that prevailed relatively quickly in neighbouring regions. Hybridisation between both spheres also constituted a recurrent trajectory. This implies a more complex process of diffusion of Corded Ware objects in central Europe.

Diverse subsistence strategies

Inherited pre-understandings also shape expectations about economic and mobility patterns, defining conceivable trajectories for members of a given society in a specific period and region (Wylie Reference Wylie2017). In this vein, many scholars have discounted foraging as a primary strategy for larger communities in central Europe once the frontier of the sedentary farming lifestyle had passed through the area. This perspective has been amplified by aDNA studies arguing for a substantial population turnover accompanying Neolithisation, effectively suggesting the physical replacement of foragers.

Even unburdened by pre-understandings, identification of late hunter-gatherers within Neolithic inland settings in central Europe is inherently challenging. Burial sites that could be associated with these groups are very rare, so the issue cannot be approached directly through bioarchaeological analyses of human remains. The identification is, therefore, dependent on the material evidence hunter-gatherers left behind, but regardless of whether researchers are equipped with conventional or state-of-the-art scientific toolkits, this is exactly where survivor bias hits. Unlike the settlement sites of sedentary farmers with their substantial domestic architecture, evidence for non-sedentary foraging communities in central Europe is mostly represented by modest surface collections of lithic artefacts, which offer limited opportunities for robust sampling, contextualisation, and even chronological resolution. Technological traditions of lithics production remained rather unchanged in many areas for centuries. Even contacts with early farmers might have had little impact on the variety of types and raw materials (see Gehlen et al. Reference Gehlen, Banghard, Schön, Otten, Kunow, Rind and Trier2015; Šída & Pokorný Reference Šída and Pokorný2020; Vondrovský et al. Reference Vondrovský, Ptáková, Šída, Bumerl, Pták, Kovačiková, Prach, Novák, Budilová and Pokorný2024 and others) or else produced a one-directional technological drift from indigenous technologies toward the traditions of incoming farmers (Gehlen Reference Gehlen, Scharl and Gehlen2017; Gronenborn Reference Gronenborn1997; Kind Reference Kind, Gronenborn and Petrasch2010; Mateiciucová Reference Mateiciucová2008, to cite but a few examples). Under such conditions, typology alone offers poor guidance. Surface scatters of chipped lithics are usually labelled ‘Mesolithic’ or, at best, ‘Late Mesolithic’. Once assigned to these categories, they are typically treated as predating the arrival of farmers or, at most, as vestiges of the relatively brief transformative phase.

However, in specific settings, particularly where a solid chronological framework can be established through radiocarbon dating of stratified deposits, a wider range of subsistence strategies becomes visible. An intriguing example comes from the Bohemian Paradise region, where long-term erosion of sandstone formations has created a unique landscape interwoven by steep, tower-like rocks and cliffs featuring countless rock shelters and caves. The stratified deposits of continuously eroding sandstone preserved inside have been systematically excavated, radiocarbon-dated, and analysed for pollen, charcoal and plant macrofossils, malacofauna, and vertebrate remains (Šída & Pokorný Reference Šída and Pokorný2020). These investigations show that the rock shelters were inhabited since the beginning of the Holocene, but the patterns and intensity remain unchanged after 5400 BCE, when early farmers settled the loess areas around the Bohemian Paradise (Ptáková et al. Reference Ptáková, Šída, Vondrovský and Pokorný2023). For the period 5400–4000 BCE, all excavated rock shelters in the Bohemian Paradise (with a single exception) testify to a continuous tradition of foraging and lithic production. Microlithic tools were chipped from locally available raw materials, underlining the distinctiveness of the area. Finds of pottery related to Neolithic cultures are sporadic and greatly outnumbered by remains of wild plants and animals, hearths, microlithic tools, or other features characteristic of Holocene hunter-gatherers in central Europe.

Around 4000 BCE, the rock shelters of the Bohemian Paradise began to be used as animal pens for domesticated livestock, primarily sheep grazing in surrounding woodlands (Ptáková et al. Reference Ptáková and Poschlod2021; Zampirolo et al. Reference Zampirolo, Holman, Sawafuji, Ptáková, Kovačiková, Šída, Pokorný, Pedersen and Walls2024). However, this transition toward forest pastoralism was not accompanied by changes in microlithic technology, which remained consistent with previous production traditions. While direct evidence on the seasonality of herding communities during the Late Neolithic is lacking, later Iron Age data indicate that livestock were overwintered in rock shelters, suggesting that the region could support year-round herding. Combined, these observations imply that the pastoral economy was maintained by relatively independent communities, while farming settlements were located on fertile soils surrounding the Bohemian Paradise or sporadically on the hillocks at its margins. Mutual connections intensified during the Early Bronze Age, which also marks the earliest pollen and plant macrofossil evidence for agriculture in the Bohemian Paradise (Pokorný et al. Reference Pokorný, Ptáková, Šída, Šída and Pokorný2020).

Lacustrine environments likewise provide stratified records revealing distinct foraging activities, as exemplified by the Třeboň Basin in south Bohemia. In the Early Holocene, this landscape featured several post-glacial lakes (Hošek et al. Reference Hošek, Prach, Křížek, Šída, Moska and Pokorný2019) around which hunter-gatherers found a suitable niche (Šída & Pokorný Reference Šída, Pokorný, Vondrovský and Chvojka2021). Švarcenberk Lake represents the best-documented case. Radiocarbon-framed paleoenvironmental data from its sediments and littoral zone document periodic burning of shoreline vegetation and a notable abundance of gathered plant species (Pokorný et al. 2010). When combined with archaeological evidence of numerous campsites around the lake (e.g. Šída et al. Reference Šída, Pokorný and Ptáková2019), a picture of an intensively exploited environment emerges. Importantly, the sequence of activities around Švarcenberk is continuous from the Early Holocene through the period following the arrival of early farmers on the nearby loess plateau after 5230 BCE (Vondrovský et al. Reference Vondrovský, Ptáková, Šída, Bumerl, Pták, Kovačiková, Prach, Novák, Budilová and Pokorný2024). Although further refinement of the radiocarbon chronology is needed, current dates suggest that intensive foraging might have continued until the mid-5th millennium BCE (Vondrovský et al. forthcoming). Despite this long temporal overlap, differences in lithic technology and raw material procurement patterns indicate limited interaction between foragers and contemporaneous farming groups.

Similar trajectories have been argued for some parts of southern Poland. The Neolithisation of what is now Poland in the 6th millennium, even though patchy and far from straightforward (Bogucki Reference Bogucki, Gron, Sørensen and Rowley-Conwy2020; Płaza Reference Płaza, Grygiel and Obst2022), shifted the continental farming frontier further north into the Polish Lowlands. By the late 54th or early 53rd century BCE, farmers reached their northernmost outposts in Kuyavia, Chełmno Land, and the Pyrzyce region (Oberc et al. Reference Oberc, Czekaj-Zastawny, Rauba-Bukowska, Grygiel and Obst2022; Whittle et al. Reference Whittle, Marciniak, Pyzel, Krueger, Lisowski, Ramsey, Dunbar, Barclay, Bayliss and Gaydarska2022). Yet groups relying on a foraging economy appear to have existed well into the 5th millennium BCE in areas south of this frontier, in Lower Silesia and Lesser Poland. Labelled ‘para-Neolithic’ (Nowak Reference Nowak2013; Reference Nowak2019), in some regions these practices must be considered a possibility even during the 3rd millennium BCE.

In Lower Silesia and Lesser Poland, late hunter-gatherer sites cluster mainly on the outskirts of loess formations exploited by farmers, particularly within periglacial lowlands and on terraces above floodplains. The Bóbr, Stobrawa-Mała Panew, and Upper Warta Basins, as well as the Upper Vistula Valley near Kraków, could be named as the largest concentrations (Masojć et al. Reference Masojć, Dreczko, Mozgała, Burdukiewicz, Cyrek, Dyczek and Szymczak2009; Nowak Reference Nowak2013; Reference Nowak2021; Nowak et al. Reference Nowak, Zając and Zakrzeńska2020; for others see Kobusiewicz & Kabaciński Reference Kobusiewicz and Kabaciński1993; Kozłowski Reference Kozłowski1989). Despite close spatial proximity, hunter-gatherer and farming sites share very few characteristics. Both feature different lithic technologies and raw material spectra (Cyrek Reference Cyrek1981; Furmanek & Majosć Reference Furmanek and Masojć2016; Nowak et al. Reference Nowak, Zając, Zakrzeńska, Borić, Antonović and Mihailović2021). In the early 4th millennium BCE, pottery appears at some hunter-gatherer campsites, but it seems related to the pottery of foraging groups in eastern Europe (Cubas & Robson Reference Cubas, Robson, Nilsson Stutz, Peyroteo Stjerna and Tõrv2025), rather than to Neolithic pottery. It should be emphasised that – aside from the stratified peat-bog sequence at Pobiel 10 (Masojć Reference Masojć, Masojć, Płonka, Ginter and Kozłowski2007) – these reconstructions are largely based on radiocarbon dates obtained from charcoal, bone, or other organic materials associated with unstratified lithic scatters and hearths. Nevertheless, the increasingly large corpus of dates documenting late foraging, including bigger data series from the sites of Glanów 3, Łykowe 1, and Mokracz 1 (Nowak et al. Reference Nowak, Zając and Zakrzeńska2020), provides robust support.

Archaeological evidence recovered during rescue excavations in Muskau Heath in northern Upper Lusatia faces similar challenges, owing to low stratification, bioturbation, and the podsolisation of local sandy sediments. Nonetheless, when integrated with paleoenvironmental proxies, it suggests similar developments. Between the Mesolithic and Middle Neolithic period, local sites show no clear change in settlement and subsistence strategies. Most are located along the riverbanks, featuring lithics and sometimes also pottery fragments dispersed across former walking horizons and accompanied by pit hearts (Liebermann Reference Liebermann, Schier, Orschiedt, Stäuble and Liebermann2021). Based solely on typological classification, Mesolithic lithics, especially Late Mesolithic forms, are widespread. Ceramics can be assigned to the Stroke-Ornamented Pottery, Funnel Beaker, Globular Amphora, and Corded Ware styles, along with unclassified types that may represent autochthonous pottery produced by hunter-gatherers (Liebermann Reference Liebermann, Schier, Orschiedt, Stäuble and Liebermann2021, 120, Abb. 4; Liebermann & Jansen Reference Liebermann and Janssen2014). Beginning with the 4th millennium, palynological profiles reflect landscape changes, likely driven by land use primarily based on livestock herding. From 2900 BCE, signs of slash-and-burn management are documented by an increase in grass pollen and elevated levels of charred particles, but the earliest cereal pollen appear only from around 2200 BCE (Dinies Reference Dinies, Schier, Orschiedt, Stäuble and Liebermann2021). Archeobotanical evidence of crop cultivation, or even the processing of potentially imported crops, is absent throughout much of the Neolithic, despite intensive sampling. These activities are attested only from the Late Bronze Age (Herbig Reference Herbig, Schier, Orschiedt, Stäuble and Liebermann2021).

A common feature linking the case studies presented above is that non-conventional subsistence strategies, which do not fit archaeological expectations for a given period and area, emerge in specific environments, different from the landscapes preferred by early farmers. Where the evidence is sufficiently robust, these environments reveal long-term continuity of hunting and gathering practices from the pre-Neolithic period well into the 5th millennium. In instances where non-conventional strategies were maintained even longer, they were often replaced by herding. Evidence for crop agriculture typically did not appear until the Bronze Age, starting at the turn of the 3rd and 2nd millennia BCE. The inhabitants of these environments adhered to specific material culture traditions, such as lithics and dwellings, with no evidence for the post-built houses otherwise common during the Neolithic. These traditions can, however, be understood as tied to subsistence practices, rather than as fully independent dimensions of material representations.

Conceptualising diversity

Reconstructing prehistory, even when grounded in robust scientific datasets, remains an interpretative enterprise (R. Chapman & Wylie Reference Chapman and Wylie2016), one that routinely patches the holes of the unknown with modern preconceptions (Frieman Reference Frieman2024). Despite growing evidence revealing substantial heterogeneity in Neolithic central Europe, some of which has been presented here, most interpretations continue to privilege large-scale, homogenising narratives centred around migrations and population turnovers. The emphasis on migrations and associated principal transformations, revived by genetic research, has shifted attention away from regional trajectories. Local forms of social, economic, and cultural differentiation are frequently overshadowed, which makes it difficult to envision central Europe as a mosaic of various areas, some operating with a certain level of autonomy or even resistance within these broader processes.

The current epistemic scaffolding – assumptions about the subjects of study, the background knowledge applied to them, and the technical resources that transform surviving materials into evidence (Wylie Reference Wylie2017) – built around the central European Neolithic is rather biased and core-centred. For understandable reasons, this scaffolding has been assembled over decades from components excavated in fertile, often loess-covered lowlands. In these landscapes, the traces of the past are much easier to convert into archaeological narratives, being abundant and well-manifested. Although the resulting edifice of archaeological knowledge may appear solid, particularly when furnished with the latest scientific insights, it is inherently one-sided and effectively diminishes the testimonies of other environments. The development and transformations of core areas were indeed dynamic, yet far from universal. The scaffolding acknowledges variation, but diversity is still commonly considered in terms of isolation by distance, so that at a continental scale peripherality is confined to frontiers and fringe zones. Reinforcing itself in a positive feedback loop, this framework has become increasingly ill-suited for recognising divergent histories of various regions, where developments could have taken entirely different paths.

Nevertheless, there is room for epistemic optimism. Moments when ‘uncomfortable’ data begin to accumulate and ultimately strain existing scaffolding to its limits are turning points that perpetuate disciplinary progress (Kuhn Reference Kuhn1962). For archaeology, Alison Wylie (Reference Wylie2017) has outlined several strategies for recalibrating entrenched frameworks of inquiry. Pursuing this route requires recontextualisation of available data, meaning that evidence must be placed within newly recognised conjunctions to underpin evidential reasoning in ways that allow acknowledging the full scale of Neolithic diversity. This underscores the importance of clearly defined theoretical concepts that guide archaeological inquiries and counterbalance the broad-brush, core-centred views (Ribeiro Reference Ribeiro2019).

The small-scale trajectories that challenge the general framework, some of which were explored in the case studies, can be conceptualised as inner peripheries. This epistemic apparatus has been adopted in historical geography (Nolte Reference Nolte1991; Reference Nolte2020), but its definition remains debated in urban studies and sociology (Copus et al. Reference Copus, Mantino and Noguera2017; De Toni et al. Reference De Toni, Vizzarri, Lasserre, Carrosio, Sallustio and Di Martino2020; Kühn Reference Kühn2015), because relationships between periphery and core in general proved to be highly contextual, scale-dependent, and must be calibrated to the particular analytical setting (Anderson Reference Anderson2000). Appropriating the concept for the Neolithic context, inner peripheries can be characterised as regions following divergent trajectories that were nonetheless tightly embedded within core landscapes. The concept thereby places dynamics previously associated largely with outer frontiers into the interior landscapes, and hybridises the core and outer periphery dichotomy. Cores and inner peripheries are not exclusive entities. They are spatially close within the patchwork of central European regions, but they differ in the degree to which their communities are integrated into polythetic networks organised around specific practices and social worlds (Figure 2). Core regions are mutually more intertwined and constituted by similar practices; inner peripheries are integrated less and remain sustained largely by their own networks, although they are not fully isolated. Archaeologically, such disconnections are visible in the different temporalities, intensities, and character of cultural transmission (Chase-Dunn & Hall Reference Chase-Dunn and Hall1993; Kardulias & Hall Reference Kardulias and Hall2008).

Figure 2.

Schematic representation of inner peripheries embedded within core regions. Core regions form an interconnected polythetic framework (solid lines), whereas the inner periphery is sustained largely by its own networks (dashed), which integrate less with the core network.

Figure 2 Long description

The Venn diagram illustrates a schematic representation of inner peripheries embedded within core regions. The core regions form an interconnected polythetic framework, depicted by solid lines, while the inner periphery is sustained largely by its own networks, shown with dashed lines. These inner networks integrate less with the core network. The diagram includes multiple overlapping and non-overlapping areas, indicating the complex relationships and interactions between the core and peripheral regions.

The idea of internal diversity is not new. Several scholars have already hypothesised enduring regional pockets of distinct ancestry (Armit & Reich Reference Armit and Reich2021; Vander Linden Reference Vander Linden2016) or combinations of diverse ancestry and subsistence repertoire. Persistent hunter-gatherer groups who retained indigenous ancestry within the area settled by farmers of Anatolian ancestry are the most mentioned example (Gehlen Reference Gehlen, Scharl and Gehlen2017; Hofmann Reference Hofmann2020; Shennan Reference Shennan2018, 151; Stäuble et al. Reference Stäuble, Liebermann, Orschiedt, Schier, Schier, Orschiedt, Stäuble and Liebermann2021). It is tempting to subsume these categories of diversity into a single geographical area, especially when genetic modelling suggests the existence of parallel, not directly competing populations linked by only limited gene flow (Rio et al. Reference Rio, Quilodrán and Currat2021; Silva et al. Reference Silva and Papageorgopoulou2022) and when palaeoecological reconstructions suggest that relatively restricted lacustrine environments, enhanced through niche construction processes, could have supported foraging groups over long periods (Holst Reference Holst, Nilsson Stutz, Peyroteo Stjerna and Tõrv2025; Jochim Reference Jochim2020; Rowley-Conwy Reference Rowley-Conwy, Nilsson Stutz, Peyroteo Stjerna and Tõrv2025). Although such combinations of ancestry and subsistence are likely in many regions, caution is warranted when applying conjunctive notions of inner peripheries, encompassing various, potentially independent aspects of human life. Admittedly, reading the genetic data rigorously, the existence of inner peripheries remains only a hypothetical construct within the dynamics of Neolithic populations. Ancient DNA sampling has, in fact, revealed divergent ancestries within core populations, but it has not been able to pinpoint their primary sources. More importantly, biological ancestry, subsistence, and material representation, to stay with the categories addressed in the case studies, can operate as distinct domains that follow their own trajectories (Furholt Reference Furholt2020). For instance, in the Blätterhöhle, the presumed pairings – foragers of Mesolithic ancestry and farmers of Anatolian origin – were only partially maintained. The epistemic power of the inner periphery concept lies in the capacity to disentangle rigid, a priori schemes that may constrain how we explore the Neolithic. This also means that there can be no single or universal trajectory for inner peripheries. They are not to replace an old broad-brush scheme with a new, slightly adjusted one. Rather, the concept aligns with what Catherine Frieman (Reference Frieman2024) terms attending to unproof, that is, opening new interpretative spaces through engagement with incomplete data and persistent unknowns. The main importance of the inner periphery concept is that it draws our attention to the unknown in places where we previously projected conventional trajectories for, amongst others, economy, material representations, settlement strategies, genetic composition, or ritual and beliefs.

The toolkit provided by the scientific revolution seems rather inefficient when tracking the diversity of inner peripheries. Most methods are oriented towards sampling human tissues, yet many regions lack conventional burial sites with well-preserved human remains. Radiocarbon dating, by contrast, has emerged as an effective approach for tracing patterns that do not follow core-derived schemes. Regrettably, despite substantial advances in precision and sampling density over the past two decades, radiocarbon data are still far from available for most sites and contexts. With the coverage currently accessible, archaeologists have merely begun to reveal unconventional regional trajectories that challenge box-like frameworks of monothetic archaeological cultures and their spatiotemporal sequences (beyond cases presented above, see Diachenko et al. Reference Diachenko and Sobkowiak-Tabaka2024; Griffiths Reference Griffiths2022; Griffiths et al. Reference Griffiths, Carlin, Edwards, Overton, Johnston and Thomas2023). Future improvements in absolute dating and the epistemic apparatus of inner peripheries have the potential to dissolve the vertical sequences into more horizontal networks of coexistence, in which categories such as ‘Mesolithic’ and ‘Neolithic’ will appear more interdigitated within central European landscapes (Lucas Reference Lucas2015b; Stäuble et al. Reference Stäuble, Liebermann, Orschiedt, Schier, Schier, Orschiedt, Stäuble and Liebermann2021). This would also allow to disentangle conjunctions like ‘Neolithic farmers’, a label which has often been used uncritically as if it were the only possible configuration for the period and which merges chronology and subsistence into a single, restrictive entity, leaving little space for Neolithic hunter-gatherers.

Between environmental determinism and human agency

Environmental factors played an important role in the emergence of inner peripheries. The case studies presented here suggest that these entities often formed in environmental niches whose settings differed markedly from wider surroundings. The fine-grained mosaic of ecozones and ecological boundaries across central Europe likely amplified this trend. Wetlands, lacustrine landscapes, rocky terrains, and uplands presented environments where adhering to agricultural strategies was more challenging and thus less efficient than in core areas. Although this might imply an environmentally deterministic reading of the inner periphery concept, I do not advocate environmentally predestined or inherited peripherality (Turner & Young Reference Turner and Young2007). Inner peripheries existed within complex socio-environmental landscapes (Dörfler et al. Reference Dörfler, Dreibrodt, Valentin Eriksen, Feeser, Groß, Hofmann, Ribeiro, Schlütz, Wieckowska-Lüth, Wild, Müller, Kirleis and Taylor2024), in which peripherality was an outcome equally constructed through human agency (Kardulias Reference Kardulias2007; Kardulias & Hall Reference Kardulias and Hall2008). As such, it could be actively reinforced or diminished by decision making or by socially and technologically derived transformations of the lived world (see Bronze Age changes in Vondrovský et al. forthcoming).

The case studies have been interpreted as evidence for parallel peripheral communities that sustained distinctive lifeways or material expressions despite neighbouring systems and influences. Where genetically distinctive populations are concerned, a high degree of social boundedness, exercised through endogamous marriage preferences and taboos, appears to be the only possibility. But can such self-aware, distinctive communities be assumed in regions that were only seasonally exploited? Muskau Heath, for example, could potentially have been visited on a seasonal basis by inhabitants of the farming settlements situated in the Dresden Basin and the central Saxon loess hills (both densely settled in the Early and Middle Neolithic; Heynowski & Reiß Reference Heynowski and Reiß2010), particularly in light of the presence of conventional ‘Neolithic pottery’. Comparable dynamics have been suggested for zones of transhumance and other forms of seasonal exploitation in the lower mountain ranges (e.g. Schwarzwald: Valde-Nowak & Kienlin Reference Valde-Nowak and Kienlin2002; Bavarian Forest: Schmotz Reference Schmotz, Chvojka, Chytráček, Gruber, Husty, Michálek, Sandner, Schmotz and Traxler2019), and likewise for areas at higher elevations in the Alps and Carpathians (e.g. Brunswig & Valde-Nowak Reference Brunswig, Valde-Nowak and Garcia-Molsosa2023; Gilck & Poschold Reference Gilck and Poschlod2019; Pelisiak Reference Pelisiak, Kadrow and Włodarczak2013; von Nicolai Reference von Nicolai2023). Even sites with evidence for more substantial settlement activities, such as in the Franconian Alb, have at times been interpreted as landscapes frequented purposefully for ritual activities by groups that otherwise represented themselves through Corded Ware materials and burial practices (Turek Reference Turek2025). Drawing on the notion of taskscapes (Ingold Reference Ingold1993), some authors have stripped such areas of the possibility to accommodate socially bounded communities, pointing to a high degree of mobility within past societies (Schultrich Reference Schultrich2025).

Nevertheless, people engaged in pasturing, hunting, or quarrying within these landscapes demonstrated distinctive communities of practice. They perpetuated specific codes of settlement and economic strategies, regardless of whether they might seasonally re-coalesce into core groups. Crucially, these patterned forms of engagement recur in inner peripheries over centuries or even millennia, in some cases spanning much of the Neolithic (e.g. rock shelters in the Bohemian Paradise). Such long-term persistence created enduring contrasts and effectively sustained two, only partly connected social worlds. Movement across these physical boundaries – from core zones into inner peripheries and back – also involved transitioning between different practices and social identities. The extent of these patterns in space and time is thus central. Even when structured by seasonal rhythms at the microscale, inner peripheries unfolded within long-term temporalities that shaped how communities navigated and reproduced their worlds.

Environments are dynamic systems, whose fluctuations continually reshape ecozones and the people inhabiting them. Seasonal micro-temporalities were therefore far from fixed and could be reconfigured in times of crisis. Contrary to common expectation, environmental shifts did not always disadvantage peripheral communities. Their effects could diverge sharply across different zones (Arponen et al. Reference Arponen, Dörfler, Feeser, Grimm, Groß, Hinz, Knitter, Müller-Scheeßel, Ott and Ribeiro2019). For example, non-agrarian groups inhabiting inner peripheries may have remained relatively unaffected by long-term droughts that severely strained farming-dependent populations in the open lowlands. In such circumstances, a turn toward year-round non-agrarian strategies within inner peripheries could represent a viable and possibly the only sustainable option. Processes shaping inner peripheries should thus be understood as ongoing negotiations in which natural dynamics and human agency acted as complementary forces (Figure 3).

Figure 3.

Inner peripheries arise through the combined effects of environment and human agency, with each factor contributing to varying degrees.

Figure 3 Long description

A line graph illustrates the relationship between environment and human agency. Two triangular shapes converge, with the top triangle labeled ‘environment’ and the bottom triangle labeled ‘human agency’. The graph suggests that inner peripheries arise through the combined effects of these two factors.