Elsevier

Quaternary Science Reviews

Volume 217, 1 August 2019, Pages 268-283
Quaternary Science Reviews

The early use of fire among Neanderthals from a zooarchaeological perspective

https://doi.org/10.1016/j.quascirev.2019.03.002Get rights and content

Highlights

  • MIS 11–9 represents a period of significant techno-cultural changes, in which the regular use of fire became evident.

  • The controlled use of fire provided several crucial advantages and generated lifestyle changes for the human groups.

  • At a zooarchaeological level the main changes were in the last stages of the processing sequence and the spatial organisation.

Abstract

Fire represented a real revolution in human lifestyles, transforming the way food was processed and leading to a new way of organising settlements and interacting socially. Yet, it is one of the most debated and controversial issues in the field of Palaeolithic archaeology. The scientific community generally proposes that the regular and controlled use of fire occurred from 400 to 300 ka onward, and that the archaeological signal became well established in sites younger than 100 ka. Thus, the chronological range between 400 and 300 ka is crucial to exploring how this phenomenon and the associated behavioural changes occurred. Here, we examine the zooarchaeological signature this process left on the faunal record, including procurement techniques and animal processing (e.g., roasting). The data are compared to information from sites without fire that are framed within the same chronological period. Our objective is to collect zooarchaeological data on the process of dependence on fire as a central element in the new human mode of adaptation.

Introduction

In the scientific literature, few subjects have attracted so much research and controversy as the domestication of fire, bringing fire home or making home the place where fire is. In fact, one of the most controversial and unresolved issues in the field of Palaeolithic archaeology is determining when hominids made the transformation from using and preserving fire to producing and controlling it at will. Once it was incorporated, fire represented a real revolution in the ways of life of prehistoric human groups, transforming the way they processed food and leading to a new way of organising settlements and interacting socially within and outside the group; so much so that the emergence of controlled use of fire has been considered, along with the production of stone tools, one of the great milestones in human technological and cognitive evolution (e.g. Wrangham and Carmody, 2010; Roebroeks and Villa, 2011; Twomey, 2014; Wrangham, 2017).

There is no doubt that fire provided some adaptive advantages. It was a constant source of light and heat, which changed the pace of natural life by artificially lengthening daylight hours. This led to the development or concentration of activities around the fire and resulted in a transformation of the occupied space. Therefore, authors like Rolland (2004) propose a direct correlation between the appearance of the controlled use of fire and the emergence of home bases. This undoubtedly favoured socialization, as hearths represented a unifying meeting point for the group, enabling greater social cohesion (Wiessner, 2014) and the development of articulate language (Gheorghiu, 2002). Within the organisation of living space, fire was also used for cleaning occupation surfaces, such as in the case of Sibudu Cave, in South Africa, where the cleaning was conducted through the sweeping out of hearths and the repeated burning of bedding (Goldberg et al., 2009).

The status of fire as a heat source was also a real benefit in inclement weather and led to the possibility of exploring and living in new territories at higher latitudes. Many authors agree that the colonization of areas outside Africa, and especially in the highest regions of Europe, was linked to the use of fire as a necessary adaptive tool for the success of European settlement in these areas (Gowlett, 2006; Gilligan, 2010; Parfitt et al., 2010; Hosfield et al., 2016; MacDonald, 2018). Obviously, northern latitudes represented settlement difficulties that were not present in other areas located further south, such as the Mediterranean basin, for example. The more northern regions of Europe have severely cold winters, shorter daylight hours in winter and a lack of edible plants during the winter leading to a diet almost totally dependent on animal resources. Some authors suggest that these difficulties were overcome by a system of seasonal settlements or settlements with adaptive and technological improvements, including the incorporation of warm clothes or the controlled use of fire (e.g. Gilligan, 2010).

Fire is also considered an element of protection against predators, providing superiority over them (e.g. Brain, 1981). Apart from providing light and warmth, fire can also burn, and therefore it can easily be used as a deterrent against external threats. According to Goudsblom (1992), mastering fire ‘represented human predominance over other mammals'.

Another of its values is that it can be considered a toxin inhibitor or neutralizer. In humans, and mammals in general, gastric acid and stomach enzymes act as powerful protective barriers against most harmful bacteria to the body. However, certain toxins and bacteria can overcome this barrier, representing a danger to human health when they are digested orally. Some examples are staphylococcus or salmonella. This is where fire's heat capacity becomes useful because most pathogens do not survive high temperatures, and the enzymes responsible for digestion can act more efficiently on cooked food. That is why roasting can have a powerful preservative effect, allowing the food range to be increased, as well as improving its nutritional value or facilitating its preservation, for example, smoked foods (e.g. Wrangham, 2009). Cooking therefore enhances nutrient digestibility and reduces diet-induced thermogenesis, thereby substantially increasing the energy gained from some foods, like meat and tubers (Carmody et al., 2011).

Heat application also allows better utilization of food, improving the digestibility of proteins and increasing the absorption of some nutrients, such as iron, and the availability of the precursor for vitamin A, β–carotene (van Boekel et al., 2010). Some studies on mice and rats empirically document how cooking food increases the bioavailability of energy from carbohydrate-rich and protein-rich foods (Carmody et al., 2011) and also from lipid-rich foods (Groopman et al., 2015). These studies observe how mice gain more weight (about 2.5% more) when fed with cooked meat than when they are fed with raw meat, even when the ratio of cooked meat is lower.

Finally, fire can also be used to modify the mechanical properties of materials such as flint, wood, bone or hides. Heat treatment provides the possibility of exploiting a wider range of raw materials by improving their quality and, in the case of stones, improves the subsequent knapping process (Brown et al., 2009; Delagnes et al., 2016; Stolarczyk and Schmidt, 2018). The preparation of hafting adhesives requires also heating to obtain a homogenous mixture (Cnuts et al., 2017; Kozowyk et al., 2017).

All of these qualities make fire a revolutionary element, which, once controlled and integrated in strategies for human adaptation, becomes, as we have seen above, a trigger for countless biological, economic and socio-cultural transformations (e.g. Rolland, 2004; Wrangham, 2009; Wrangham and Carmody, 2010; Twomey, 2014). That is why determining when this revolution took place and the process that led to its implementation is crucial to understanding our own evolutionary history.

Most archaeological evidence for early use of fire is based on the macroscopic identification of burnt items, which are usually described by changes in colour, and sometimes by alterations in physical structure, such as cracks or fissures (e.g. Shipman et al., 1984). However, in some cases, the appearance of burning may be misleading since a blackish colour with a uniform appearance could also correspond to alterations from manganese oxides. The blackish pigmentation characteristic of manganese is produced by bacteria that thrive in moist and aerobic environments with a pH that is close to neutral, but it is also typical of environments characterised by alternating cycles of oxidation-reduction (Fernández-Jalvo and Andrews, 2000).

Proof of how easy it is to confuse this alteration with the damage caused by thermal action is the famous case of Zhoukoudian, dated between roughly 700 and 200 ka (Goldberg et al., 2001; Zhong et al., 2014; Gao et al., 2017). For decades, the so-called Peking Man (Homo erectus) was considered to be a hominid that engaged in the controlled production and management of fire; however, subsequent analyses have cast doubt on this assertion. Not only were the blackish colourations due to manganese, but also the use of soil micromorphology and Fourier transform infrared spectrometry (FTIR) analyses of the sediments showed the so-called hearths from this locality as laminated waterlain silts and organic matter accumulations (water-laid accumulations), all of them transported (Goldberg et al., 2001). Given the difficulty and the interpretative implications, Zhong et al. (2014) also resorted to the use of other techniques to seek the microscopic criteria of heat alteration, such as advanced X-ray diffraction. Although this analysis supports the presence of burnt bones associated with burnt flint at Zhoukoudian, these remains are not associated with in situ anthropogenic combustion features. However, a recent study indicates that Layer 4 (the upper cultural horizon) contains clear-cut evidence for in situ use of fire using measurements of magnetic susceptibility, colour, and diffuse reflectance spectra of sediments (Gao et al., 2017). In any case, Zhoukoudian exemplifies the complexity of identifying hearths in ancient contexts and the growing need to incorporate micromorphological, petrographic and radiometric techniques, as well as the development of spatial distribution analysis (e.g. Roebroeks and Villa, 2011).

Apart from the complexity of identifying the thermal impact, one of the great challenges that the researchers face is that sometimes it is difficult to separate in the archaeological record the evidence for fire as a result of human action from that generated by natural fire, especially in the case of other processes that may also conceal the actual sedimentary thermal alteration generated by a fireplace (Preece et al., 2007). For example, Hendey (1976) reported burnt bones from the fossil bone bed of the Miocene horizon of Langebaanweg.

In addition, identification of human use of fire is further complicated by the possible human exploitation of natural fire. Natural fires have always been part of the process of stabilization and transformation of ecosystems (Pacault, 1995), but the frequency with which they occur is more random, sporadic and, above all, restricted to certain environments. Some authors describe various ways in which fire occurs naturally (e.g. Perles, 1977): sparks from rock falls, volcanic eruptions (although their frequency would be limited to regions of volcanism), spontaneous combustion of coal, fermentation of decaying plant matter (due to the production of certain gases, such as hydrogen phosphide or methane), meteorite impacts and, mainly, lightning strikes, which are still the main and most important cause of spontaneous fires in nature today (although the generation and spread of fire depends on the density of vegetation and its state of dryness).

There are multiple ways in which a natural fire can occur, but it tends to remain in certain geographical areas or environments, so the probability that a group of early humans would find a fire was not very high. This reflection, according to Perles (1977), has two working hypotheses: 1) the production of fire was immediate because of it was difficult to obtain through forest fires and hominids would have found out how to produce it when they began using it; 2) the production of fire was preceded by a long period of previous experiences, the first of which was learning how to maintain and conserve it.

These options are not necessarily contradictory, although we do generally tend to distinguish three stages in the course of the conquest of fire (Stahlschmidt et al., 2015, and references therein): the use (including both the conceptualization of fire and its collection from natural sources), control (maintenance of a fire via fuel provisioning and restraint) and production (artificially produced by wood-on-wood friction or stone-on-stone percussion in addition to a tinder source).

But perhaps the step that triggered the greatest revolution was not exclusively the generation of fire, but the step before, with the transition from ‘non-use’ to ‘use’ (control), because it was at this point that hominids began to discover its benefits and it became a valuable item for livelihood strategies, thus producing a real mental revolution to generate it (Perles, 1977; Gómez de la Rua and Díez, 2009). For other authors, however, the crucial phase is still production, as that is when innovation and a generalization of the associated changes truly occur. These researchers even propose that fire use (without production) may not be exclusive to the human species, as some actualist observations show how some animals are capable of exploiting wild fires for warmth and food (Wrangham, 2009).

Human knowledge of fire is difficult to identify in the archaeological record and most hypotheses concerning early hominid–fire interaction are based on inferences from indirect data. An example that illustrates this situation is how Gowlett and Wrangham (2013) used the changes in human anatomical morphology to defend the use of fire as early as 1.5 Ma, despite the lack of direct archaeological evidence.

Some authors hypothesise that fire use originated in Africa, at sites such as GnJi 1/6 in Chesowanja, Kenya (1.42 Ma), where thermo-altered clays were identified. The chemical and magnetic analysis conducted at the site reported temperatures around 400 °C on altered sediments. These data were considered important since natural fires (both trees and pastures) do not usually exceed 250 °C (Gowlett et al., 1981). However, James (1989) advised that, although these temperatures could correspond to controlled heat, natural combustion of bushes cannot be totally ruled out.

The site of FxJj 20M in Koobi Fora (∼1.6 Ma), Kenya, where oxidized sediments with a thickness of less than 50 cm were recorded, follows the same pattern. Bellomo (1994) conducted a review of the archaeological evidence, including spatial studies, and determined that although nothing could be linked to cooking activities, at least one of the hearths of FxJj 20M could have been a source of light and heat for actions such as lithic reduction or for use as protection against predators. Linked to this, a recent study focused on the nearby site of FxJj20 AB (∼1.5 Ma) supports evidence of thermally altered lithics, soil aggregates and bone fragments using FTIR analyses, as well as a spatial pattern consistent with prehistoric anthropogenic fire features from Eurasian archaeological sites (Hlubik et al., 2017).

Another site where the presence of fire is controversial is Member 3 (about 1.5–1.0 Ma) from Swartkrans Cave, South Africa, where 270 burnt bones from the hominid-bearing breccias were recovered (Brain and Sillen, 1988). The histology and chemistry of the altered items indicated that they were heated to a range of temperatures consistent with that occurring in campfires (300–500 °C). According to Brain and Sillen (1988), the fact no burnt items have been recorded in other sedimentary units (Members 1–2) of the site, despite having similar environmental conditions, could be an argument in favour of the intentional use of fire. However, the authors also warned that this evidence does not necessarily prove the fire was used for cooking because protection against predators and the provision of warmth are equally plausible uses.

Other controversial evidence also includes the 8E site of Gadeb (1.45–0.7 Ma) in Ethiopia. At this site, a group of stones that appeared to present signs of heat alteration was recorded. In this case, palaeomagnetic analyses were performed but failed to show whether the heat-altered rocks were formed through human actions or volcanic activities in the area (James, 1989).

Some of the earliest evidence for knowledge of fire in Africa, which has not yet been refuted, comes from the ∼1 Ma site of Wonderwerk, South Africa, where heated sediments, ash remains and burnt bones, identified by micromorphological and mFTIR analysis, were found inside a cave. This fact was used by Berna et al. (2012) to suggest that this phenomenon could not be explained by a natural fire. However, this behaviour does not seem to have continuity in Africa until well into the Middle Pleistocene. In some cases, such as Bodo A4 (0.65 Ma) and HAR A3 (0.5 Ma) in Awash, Ethiopia, heat-altered clays 40–80 cm in diameter have been documented, together with various stone tools and even cranial remains of Homo rhodesiensis (BOD-VP-1/1). However, James (1989) identified some problems with these sites; for example, there is no proof that burning at these locations was not fortuitous, and there is no explanation for the lack of burnt wildlife. Other evidence of the use of fire in Africa during this period has been suggested in the Cave of Hearths (∼0.2 Ma), Klasies River Mouth (∼0.12 Ma) and Montagu Cave (∼0.12 Ma) in South Africa; in Member 7 of Olorgesailie (∼0.4 Ma), Kenya; and in Kalambo Falls (∼0.12 Ma), Zambia (James, 1989). It is worth highlighting the case of the Cave of Hearths, as it exemplifies how the use of fire has been widely questioned, and explanations related to natural phenomena or post-depositional alterations have been proposed. When it was discovered, the Cave of Hearths was considered the earliest site of the Middle Pleistocene with evidence of fire, due to its thermo-altered areas and concentrations of small altered bones (Oakley, 1955, 1956); however, further analysis showed that the alterations were actually bat guano (James, 1989; see a review in Díez Martín, 2005).

In Asia, early evidence of fire is even more problematic and biased, if possible. An example is the site of Yuanmou in China where several charcoals and two bones with blackish colouring were initially interpreted as burnt. The site presents chronology problems, with palaeomagnetic dating ranging from 1.7 Ma to 600–500 ka (James, 1989, and references therein). Another case with similar problems is the Xihoudu site, also in China. As in Yuanmou, bones were recovered with blackish colouring, and palaeomagnetic dating lies between 1.8 and 1 Ma (James, 1989). The site of Zhoukhoudian, with a chronology from 700 to 200 ka, was once considered the earliest site with clear controlled use of fire. However, as discussed above, the site has not been free of controversy (e.g., Oakley, 1956; Goldberg et al., 2001; Zhong et al., 2014; Gao et al., 2017). Other major Asian sites for the Middle Pleistocene are Gongwangling and Jinniushan, also in China (Keates, 2000), and Trinil, in Java (Carthaus, 1911). Only Jinniushan, dated to ca. 200 ka by Electron Spin Resonance (ESR) and Uranium-series (U-series), shows evidence of fire on the appearance of burnt materials, which has been confirmed by chemical analysis (James, 1989).

In the Middle East, the oldest evidence of intentional fire appears to be located at the site of Gesher Benot Ya'aqov, Israel, dated to between 0.7 and 0.8 Ma (Goren-Inbar et al., 2004). Although micromorphological features have not been described for this site, Alperson-Afil et al. (2007, 2009) supported the presence of hearths from analyses based on the spatial distribution of flint artefacts using geographic information systems (GIS) and thermoluminescence analyses. However, in their discussion of burning by natural fire, they only refuted the possibility of an in situ natural fire; they did not discuss the possibility that raw material had been heat-altered before hominid use. In any case, it is worth noting the curious appearance of six burnt wood and herbaceous taxa –three of which belong to edible species (olive, barley and grape)– in association with the anthropogenic materials in the site (Goren-Inbar et al., 2004).

Within the European context, there is a recently published case of old fire identified at Cueva Negra del Estrecho del Río Quípar, Spain, with dates between 780 and 500 ka, determined by magnetostratigraphy, Optically Stimulated Luminescence (OSL) and micro-mammalian biochronology (Walker et al., 2016). TL, FTIR, Electron Microscopy and taphonomical analysis of bone fragments attributed discolouration to burning, not to post-depositional mineral staining. However, sediment geochemistry and thin-section micromorphology analyses have not been carried out to confirm or reject the presence of in situ anthropogenic combustion structures.

Despite the isolated case of Cueva Negra del Estrecho del Río Quípar, the scientific community generally suggests that regular and controlled use of fire occurred from 400 to 300 ka onward and that the archaeological signal becomes common and well established in sites younger than 100 ka (Roebroeks and Villa, 2011). According to Gowlett (2006), fire is absent in Europe before the Anglian Glaciation (MIS 12). Below, we will mention the most significant sites for their antiquity and preservation, as our intention in this work is to address the timing and manner in which fire was introduced in the faunal processing sequence.

Perhaps the earliest, most direct and unambiguous evidence for the control and habitual use of fire in the form of hearths and reused hearths comes from the Near East, at Qesem Cave in Israel, dated in the lower sequence to 400 and 300 ka. At this site, a combustion feature containing two superimposed use cycles with a chronology of 300 ka was reported (Shahack-Gross et al., 2014). The combustion feature is grey-coloured, covers an area of approximately 4 m2 and is located in the central part of the cave, away from the cave walls. The area around the hearth bears dense faunal and lithic remains, as well as evidence for spatial differentiation of activities (Blasco et al., 2016). The lowest levels, dated to >400 ka, produced similar proportions of bones and heat-altered lithics, suggestive of regular use of fire throughout the entire sedimentary sequence (Barkai et al., 2017). Although the Tabun Cave archaeological site (350 ka) in Israel does not present hearths or burned faunal remains, this locality yields high frequencies of burned flints that seem to support the use of fire in the Levant during these chronologies (Shimelmitz et al., 2014).

Burnt material at the archaeological sites of Vertesszöllös in Hungry (Kretzoi and Dobosi, 1990), Menez-Dregan (Monnier et al., 2005) and Terra Amata in France (Lumley, 2016), Bilzingsleben in Germany (Mania and Mania, 2000, Mania et al., 2005), Beeches Pit in England (Gowlett, 2006; Preece et al., 2006), Maastricht-Belvédère in the Netherlands (Roebroeks, 1988) and Bolomor Cave in Spain (Fernández Peris et al., 2012) have often been reported as the earliest evidence of human control of fire in Europe. However, as with the other locations mentioned above, some researchers have warned of different problems relating to the chronological allocation and the taphonomic processes of some of these sites.

Good examples of these dating problems include the site of Menez-Dregan on the Atlantic coast of France (Vliet-Lanöe and Laurent, 1996) and Vertesszöllös in Hungary (Kretzoi and Dobosi, 1990). Menez-Dregan I presents a series of structures that could correspond to fireplaces associated with charcoals and burnt tools on levels 9, 7, 6 and 5, which have been dated by ESR to around 465 and 380 ka (Monnier et al., 2016). Despite evidence of heat alteration, TL dates on heated quartz and flint grains have given much younger ages (Mercier et al., 2004). Vliet-Lanöe and Laurent (1996) warned of a possible alteration in the datings due to radioactivity in the granitic composition of the sediments. Nevertheless, Monnier et al. (2016) have retained the ESR dates as the most plausible, considering the geological context and archaeological data. Issues with chronology also occur at Vertesszöllös. The (U-series) absolute chronology of travertine formations between 185 and 211 ka, is considered too late for the micro-vertebrates represented at the site, with the presence of Arvicola terrestris cantiana, and therefore, it should have an age closer to 350 ka. Apart from the chronology, James (1989) also argued that the areas recorded as heat-altered could correspond to mineral stains caused by low temperatures and humidity.

A different problem exists for Terra Amata, located on the Mediterranean coast of Nice, France, where the in situ record has been questioned as a result of a series of lithic remounts that demonstrate alteration processes and remobilization of sediments (Villa, 1982; Gamble, 2001). Excavations carried out in the sixties revealed several combustion structures (see review in Lumley, 2016). At first, this site gave a chronology between 250 and 400 ka, but later, two burnt flints were analysed using TL, yielding a date of 230 ka. However, the mammal assemblage characterizes a warm period of the Middle Pleistocene, attributed to the early Aurelian formation (MIS 9 or 11) (Valensi et al., 2011). Besides burnt flints, several charcoals (Pinus sylvestris) and burnt bones have also been located.

The site of Bilzingsleben, Germany, has been dated between the chronological lapses of 350–320 ka and 414–280 ka (Mania and Mania, 2000; Gamble, 2001) by Th230single bondU234 and ESR. Here, accumulations of burnt remains forming semi-circular areas were recovered, but with no evidence of sediment heat alteration (Mania and Mania, 2000). Despite the presence of fire, it has not yet been demonstrated whether the evidence of fire was produced by human activity or natural processes. Also, in Germany, the Schöningen site was referenced as among the earliest examples of controlled use of fire. Recently, this hypothesis has been refuted, and Stahlschmidt et al. (2015) concluded that the analysed features and artefacts present no convincing evidence for human use or control of fire. This reassessment was carried out through a multianalytical, micro-contextual approach including techniques such as micromorphology, FTIR, organic petrology, luminescence and analysis of mineral magnetic parameters.

Beeches Pit, in England, also shows fire activity in the form of flint, burnt bones and rubifacted areas of ∼1 m2, which have been interpreted as hearths (Gowlett, 2006; Preece et al., 2006, 2007). The site has been dated by TL, U-series and Amino Acid Racemization (AAR), giving an age of around 400 ka. However, four OSL dates from sand beneath the interglacial sequence yielded a mean age of 200 ka, far younger than all other age determinations and than implied by the biostratigraphy (Preece et al., 2007). According to Preece et al. (2006), there are several indications that rule out a natural fire at the site for the following reasons: 1) the areas of combustion coincide stratigraphically with the largest accumulations of material in the sequence; 2) the rubifacted areas are restricted to shallow depressions; and 3) the spatial distribution of the materials surrounds the burnt areas. For these authors the presence of charred bones (in grey or white shades) implies more intense combustion than that usually caused by natural fires. The use of X-ray diffraction (XRD) to analyse three bones from Bed 6 confirmed that the charred and calcined bones were intensely heated (Preece et al., 2007). However, Stahlschmidt et al. (2015, p.182) considered this data insufficient to demonstrate that the supposed hearths were formed through heating, and it is unclear if the heated bones and lithic artefacts were produced by human activities or natural fires.

Another site that deserves mention is Maastricht-Belvédère, in the Netherlands (Roebroeks, 1988; Stapert, 1992). Radiometric techniques used at the site included TL dating of heated flint artefacts, which yielded an age of 250 ka, and ESR dating of shells, which yielded an age of 220 ka. At this site, several heat-altered fragments were located, concentrated into two main groups. Stapert (1992) observed a unimodal distribution, separate from the centre of the feature where the items appeared with signs of having been burnt, suggesting a possible structure for open-air combustion. By contrast, Roebroeks (1988) argued that the origin of this concentration could be ascribed to natural causes, an interpretation that appears to be corroborated by geological studies of the site.

One of the best-conserved sites is Bolomor Cave in Spain. This site has provided an important chrono-stratigraphy for the Middle Pleistocene, with dates ranging from 350 to 100 ka (Fernández Peris et al., 2012). Although the heat-altered material has been recovered at the lowest level of the sequence (XVII, 350 ka), the oldest combustion structures come from level XIII, with an age of 230 ka determined by AAR. The two hearths documented at this level have a complex structure; one of them is basin-shaped and the other shows preparation prior to ignition in the form of stone beds to insulate it from the ground. At level XI, ∼150 ka, seven simple oval-shaped hearths were documented, which were aligned under the start of the cave's ledge. Around the hearths, a significant accumulation of archaeological material was documented. Levels II and IV, with chronologies of 120–100 ka, have also provided evidence of controlled use of fire. In level II, only ash accumulations were recorded, while in level IV, four hearths were documented, also located under the line of the overhang, on the west side of the cave mouth.

Lastly, owing to its similar chronology, it is also worth highlighting the site of Cotte de Saint Brelade, on the island of Jersey, UK, where abundant cremation remains were recorded in layers C–D, with a chronology around 230 ka. Despite recovering heat-altered archaeological remains, there was no evidence of combustion structures (Callow and Cornford, 1986).

As illustrated by the sites listed above, evidence of fire in its early stages (Early/Middle Pleistocene) of use by humans has always been a controversial subject, not only due to the issues involved in its identification in early contexts, but also due to difficulties in understanding how human groups implemented it in their domestic and daily activities. At the zooarchaeological level, significant changes would be expected, including animal procurement techniques, processing and consumption. Thus our goal is to explore these changes in European Neanderthal sites with an early presence of fire around 400–300 ka and to compare them with contemporary deposits where the use of fire has not been documented.

Section snippets

What does the zooarchaeology say?

Although there are several Middle Pleistocene sites with evidence of fire in Europe and the Levant, there is limited zooarchaeological information from those deposits framed within MIS 11–9. Here, only the sites with clear stratigraphical and chronological control and recent taphonomic studies are considered (see Table 1, Fig. 1).

Discussion and conclusions

The MIS 11–9 entails a high diversity of archaeological sites that present both different physical and functional characteristics. On the one hand, it is possible to distinguish between sites located in karstic environments (caves and shelters) and those located in open-air areas, such as lake edges, river terraces and/or colluvial deposits. On the other hand, a certain variability of activities is also identified, which translates into different archaeological interpretations. In this section,

Acknowledgements

J. Rosell and R. Blasco develop their work within the Spanish MINECO/FEDER projects CGL2015-65387-C3-1-P, CGL2016-80000-P and CGL2015-68604-P, and the Generalitat de Catalunya projects 2017 SGR 836 and CLT009/18/00055. We sincerely thank J.S. Carrión for allowing us to participate in this special issue.

References (167)

  • J. Fernández Peris

    The earliest evidence of hearths in southern Europe: the case of Bolomor cave (Valencia, Spain)

    Quat. Int.

    (2012)
  • S. Gaudzinski et al.

    Palaeoecology and archaeology of the Kärlich-Seeufer open-air site (middle pleistocene) in the central Rhineland, Germany

    Quat. Res.

    (1996)
  • P. Goldberg

    Site formation processes at Zhoukoudian, China

    J. Hum. Evol.

    (2001)
  • J.A.J. Gowlett

    The early settlement of northern Europe: fire history in the context of climate change and the social brain

    C R Palevol

    (2006)
  • J.F. Hoffecker et al.

    Large mammal taphonomy of the middle pleistocene hominid occupation at Treugol’naya cave (northern Caucasus)

    Quat. Sci. Rev.

    (2003)
  • R.G. Klein

    Why does skeletal part representation differ between smaller and larger bovids at Klasies River Mouth and other archaeological sites?

    J. Archaeol. Sci.

    (1989)
  • A.M. Moigne et al.

    Les faunes de grands mammifères de la Caune de l'Arago (Tautavel) dans le cadre biochronologique des faunes du Pléistocène moyen italien

    L'Anthropologie

    (2006)
  • A.M. Moigne et al.

    Bone retouchers from Lower Palaeolithic sites: Terra Amata, Orgnac 3, Cagny-l'Epinette and Cueva del Angel

    Quat. Int.

    (2016)
  • M.H. Moncel et al.

    Pre-Neandertal behaviour during isotopic stage 9 and the beginning of stage 8. New data concerning fauna and lithics in the different occupation levels of Orgnac 3 (Ardèche, South-East France): occupation types

    J. Archaeol. Sci.

    (2005)
  • M.H. Moncel et al.

    Towards the middle Palaeolithic in western Europe: the case of Orgnac 3 (southeastern France)

    J. Hum. Evol.

    (2012)
  • J.-L. Monnier et al.

    Menez-Dregan 1 (Plouhinec, Finistère, France): un site d’habitat du Paléolithique inférieur en grotte marine. Stratigraphie, structures de combustion, industries riches en galets aménagés

    L’anthropologie

    (2016)
  • A. Agam et al.

    Elephant and mammoth hunting during the Paleolithic: a review of the relevant archaeological, ethnographic and ethno-historical records

    Quaternary

    (2018)
  • N. Alperson-Afil

    Phantom hearths and the use of fire at Gesher Benot Ya’akov, Israel

    Palaeoanthropol

    (2007)
  • N. Alperson-Afil et al.

    Spatial organization of hominin activities at Gesher Benot Ya‘aqov, Israel

    Science

    (2009)
  • A.P. Anzidei et al.

    Le gisement pléistocène de La Poledrara di Cecanibbio (Rome, Italie)

    L'Anthropologie

    (1989)
  • A.P. Anzidei

    Tools from elephant bones at La Polledrara di Cecanibbio and Rebibbia-Casal de'Pazzi

  • A.P. Anzidei et al.

    The fauna of La Polledrara di Cecanibbio and Rebibbia-Casal de’ Pazzi (Rome, Italy) as an indicator for site formation processes in a fluvial environment

  • M. Arilla et al.

    Contributing to characterise wild predator behaviour: consumption pattern, spatial distribution and bone damage on ungulate carcasses consumed by red fox (Vulpes vulpes)

    Archaeological and Anthropological Sciences

    (2018)
  • R. Barkai et al.

    Fire for a reason. Barbecue at middle pleistocene Qesem cave, Israel

    Curr. Anthropol.

    (2017)
  • L.E. Bartram

    Perspectives on skeletal part profiles and utility curves from Eastern Kalahari Ethnoarchaeology

  • F. Berna

    Microstratigraphic evidence of in situ fire in the Acheulean strata of Wonderwerk cave, northern Cape province, South Africa

    Proc. Natl. Acad. Sci. Unit. States Am.

    (2012)
  • L.R. Binford

    Nunamiut Ethnoarchaeology

    (1978)
  • L.R. Binford
  • R. Blasco et al.

    Subsistence economy and social life: a zooarchaeological view from the 300 kya central hearth at Qesem Cave, Israel

    J. Anthropol. Archaeol.

    (2014)
  • E. Boëda

    Approche de la variabilité des systémes de production lithique des industries du Paléolithique inférieur et moyen

    Chronique d'une variabilité attendue. Techniques et cultures

    (1991)
  • G. Boschian et al.

    In the elephant, everything is good: carcass use and re-use at Castel di Guido (Italy)

    Quat. Int.

    (2014)
  • C.K. Brain

    The Hunters or the Hunted?

    (1981)
  • C.K. Brain et al.

    Evidence from the Swartkrans cave for the earliest use of fire

    Nature

    (1988)
  • K.S. Brown

    Fire as an engineering tool of early modern humans

    Science

    (2009)
  • I. Cáceres et al.

    Fresh and heated bones breakage: an experimental approach

  • E. Carthaus

    Spuren von möglicherweise menschlicher Tätigkeit in den Trinil- Schichten

  • R.N. Carmody

    Energetic consequences of thermal and nonthermal food processing

    Proc. Natl. Acad. Sci. Unit. States Am.

    (2011)
  • P.F. Cassoli et al.

    Giacimento del Paleolitico inferiore a Malagrotta (Roma). Atti 33a Riunione Scientifica Istituto Italiano Preistoria Protostoria, Firenze

    (1982)
  • D. Cnuts et al.

    The role of fire in the life of an adhesive

    J. Archaeol. Method Theory

    (2017)
  • N.J. Conard et al.

    Middle Palaeolithic hunting economies in the Rhineland

    Int. J. Osteoarchaeol.

    (2000)
  • S. Costamagno et al.

    Approche expérimentale d’un problème taphonomique : utilisation de combustible osseux au Paléolithique

    Préhistoire Européenne

    (1999)
  • S. Costamagno et al.

    Taphonomic consequences of the use of bones as fuel : experimental data and archaeological applications

  • S. Costamagno

    Bone grease rendering in Mousterian contexts: the case of Noisetier cave (Fréchet-Aure, Hautes-Pyrénées, France)

  • S. Costamagno et al.

    L’exploitation de la graise au Paléolitique

  • Cited by (13)

    • Fire among Neanderthals

      2022, Updating Neanderthals: Understanding Behavioural Complexity in the Late Middle Palaeolithic
    • Late Acheulian multiplicity in manufactured stone culture at the end of the Middle Pleistocene in Western Europe

      2021, Quaternary International
      Citation Excerpt :

      Such, home-base provisioned occupations became even more pervasive from the Lower-Middle Paleolithic transition during the Upper Pleistocene (esp. after 200- 100 Ka) (Valensi et al., 2013; Rosell and Blasco, 2019; Vallverdú et al., 2012) (Fig. 6). While it is convenient to establish a relationship between the presence of hearths and the increase of rock-shelter and cave occupations, perceived in the archeological register, this postulate may indeed be biased by geological factors, which, in turn, depend on the preservation of fire indicators (charcoal, carbonized rock or sediment, charred artifacts) and our ability to detect them in different site contexts, a condition that is understandably more likely to occur in archeologically controlled enclosed cave settings (Roebrooks and Villa, 2011).

    • Shivering in the Pleistocene. Human adaptations to cold exposure in Western Europe from MIS 14 to MIS 11

      2021, Journal of Human Evolution
      Citation Excerpt :

      Bilzingsleben and Schöningen, in Germany, have also been claimed to provide evidence of fire control by humans at around MIS 9 or MIS 11, but this interpretation has been questioned (Stahlschmidt et al., 2015; Rosell and Blasco, 2019). Rosell and Blasco (2019) suggest that fire control and the emergence of post-Acheulean technology occurred in Europe in parallel, between MIS 11 and MIS 9. In summary, evidence of fire is extremely rare before MIS 9, and it increases progressively to become frequent during MIS 4 and MIS 3 (Roebroeks and Villa, 2011).

    • The metapodial hammers from the Lower Palaeolithic site of Schöningen 13 II-4 (Germany): The results of experimental research

      2021, Journal of Archaeological Science: Reports
      Citation Excerpt :

      The rare occurrence of metapodial hammers could be related to opportunistic behaviour applied under very specific circumstances; behaviour that is not used often or widespread, nor well embedded in the material- or cultural tradition. Technological and cultural innovations took place during the transition from the Lower to the Middle Palaeolithic in Europe and the Near-East (Hopkinson, 2007; Roebroeks and Villa, 2011; Fontana et al., 2013; Picin et al., 2013; Adler et al., 2014; Di Modica and Pirson, 2016; Malinsky-Buller, 2016; Rosell and Blasco, 2019) and Schöningen has evidence of this transition as well. While its lithic assemblage has been assigned to the Lower Palaeolithic due to the absence of Levallois technology (Serangeli and Conard, 2015), the site does have some indications for technological novelties.

    • Neanderthals: Ecology and evolution

      2019, Quaternary Science Reviews
    View all citing articles on Scopus
    View full text