Journal of Paleolithic Archaeology

, Volume 1, Issue 1, pp 54–79 | Cite as

Were Western European Neandertals Able to Make Fire?

  • Harold L. DibbleEmail author
  • Dennis Sandgathe
  • Paul Goldberg
  • Shannon McPherron
  • Vera Aldeias


Significant variability has been observed in the frequency of fire use over the course of the Late Pleistocene at several Middle Paleolithic sites in southwest France. In particular, Neandertals appear to have used fire more frequently during warm climatic periods and very infrequently during cold periods. After reviewing several lines of evidence and alternative explanations for this variability, the null hypothesis that these Neandertals were not able to make fire still stands.


Pyrotechnology Paleolithic fire Middle Paleolithic Western European Neandertals 


Based initially on our work at two Middle Paleolithic sites in southwestern France, Pech de l’Azé IV (Pech IV) and Roc de Marsal, a pattern emerged that suggested varying use of fire by the sites’ occupants (Sandgathe et al. 2011; Aldeias et al. 2012; Goldberg et al. 2012; Dibble et al. 2017). Variation in fire use would not be particularly surprising in itself—of course there are a multitude of reasons why people did or did not use fire at any particular time and place. What was surprising, however, is that at both of these sites, there was ample evidence for fire use in layers occupied during warm periods, but during colder periods evidence shows that fire use dropped dramatically over long stretches of time. Given the importance of pyrotechnology to human groups in recent times, seeing long-scale reduction in its use—especially during cold periods—is counter-intuitive and, therefore, demands some explanation. One possibility we suggested was that it reflected a lack of technical knowledge of how to start fires, and that western European Neandertals instead relied on access to natural fires (Sandgathe et al. 2011). This hypothesis is based on the premise that lightning occurs more frequently during warmer periods and, as it is the primary cause of natural fires, there would therefore be more natural fires available to exploit during warm periods and fewer during cold phases.

Following subsequent discussions with other researchers and in response to counter-arguments made in a recent publication (Sorensen 2017a), we present here additional data, arguments, and clarification relevant to our hypothesis. There are numerous factors—both natural and human—that potentially influence variability in fire use, and it is clear that this is a complex problem that involves several different lines of evidence. There are two issues, however, that should be kept separate. One concerns the question of fire-making abilities. Our hypothesis that Neandertals harvested naturally occurring fire is one potential explanation for the observed patterning, but there are others, and some of these will be discussed below. The other issue concerns the patterns themselves: if fire use was reduced during the cold periods, then how did those hominins adapt to such extreme conditions? The answer to this question will require much more data from a larger number of sites, and, more specifically, a greater understanding of the purposes for which they were using fires and further investigation of alternative technologies, behaviors, or physiological changes that could ameliorate the situations in which they were living.

One model for the evolution of fire-related behaviors outlines several steps, starting with habituation to natural fires, through the simple use of fire for specific purposes, the ability to control and maintain them, and, ultimately, to the ability to start them (Pruetz and Herzog 2017; Pruetz and LaDuke 2010; Sandgathe 2017). Documenting the first of these in the hominin line is likely to be impossible, and while there is suggested evidence that the origin of human fire use extends back to more than 1 Ma (Gowlett and Wrangham 2013; Hlubik et al. 2017; Rowlett 2000; Berna et al. 2012; Barbetti 1986; and see Sandgathe and Berna 2017), there is little doubt that the regular use and maintenance of fire by hominins is present in Middle Paleolithic and Middle Stone Age contexts and later (Roebroeks and Villa 2011; Barkai et al. 2017; Shimelmitz et al. 2014; Aldeias et al. 2014; Dibble et al. 2009, 2017; Goldberg et al. 2012; Rodríguez-Cintas and Cabanes 2017; Mallol et al. 2013; Miller 2015; Courty et al. 2012; Pasto et al. 2000; Vallverdú et al. 2012; Pop et al. 2016; Vidal-Matutano 2016). At some point, hominins also developed the technology for starting fires, which is the question being addressed here.

It should be emphasized that throughout this paper we are referring to Middle Paleolithic hominin populations in a portion of Western Europe. Technological knowledge—including pyrotechnology—may have varied considerably among Neandertal populations. It is also important to emphasize that it is logically impossible to prove that Neandertals—or any other hominin—lacked the knowledge of how to start fire. Such a conclusion would, essentially, be an argument from ignorance, which is a logical fallacy in the same sense that an absence of evidence is not necessarily evidence of absence. However, the ability to ignite fires is a technological innovation that, like all such innovations including the ability to control and maintain fire, occurred at some point in the hominin lineage. As archaeologists, our goal is to document when and where specific technological developments first occurred and how they spread, but we also have to accept that the null hypothesis must be that a particular technology was not present until such documentation is made. Therefore, logical reasoning dictates that we view the inability to make fire as the default position until a stronger alternative explanation can better account for the evidence we have.

The null hypothesis that Neandertals, at least in southwest France, lacked an ability to make fire can be tested and refuted. One way to refute it would be to demonstrate that the frequency of natural fires is not related to climate, since our hypothesis rests on that assumption. If climate is unrelated to the occurrence of natural fires, then it implies that changes in fire frequencies seen in Middle Paleolithic sites must be due to some other factor. Another approach is to expand the investigation to other sites. If the patterns we described do not hold up, for example that other sites show abundant evidence for fire during cold periods, then we would conclude that regional climatic variation is not affecting fire frequencies. A third possibility is that the pattern continues to hold at the regional level, but other alternative explanations better explain the patterns we see. Of course, such alternatives would also have to account for variation that occurs in relation to climate. While some of these alternative explanations may or may not be sufficient to reject the null hypothesis itself, if one or more of them fit better with other known data our attention could shift to test them further. And finally, perhaps the strongest refutation of our hypothesis would be the discovery of artifacts, features, or other evidence that unambiguously attests to fire-making technology. In the following pages, these possible refutations or alternative explanations of the null hypothesis will be discussed in turn.

Do Lightning-Caused Wildfires Vary with Climate?

In our papers (see especially Sangathe et al. 2011), we noted that the major cause of natural fires is lightning and that lightning is more frequent during warmer and more humid conditions. At the time of our original publication, we did not feel it was necessary to document extensively the associations between climate, lightning, and wildfires—they are well-known and well-understood meteorological phenomena that operate over both the short term—daily or seasonal (weather)—and the long term (climate), primarily as a response to variation in solar radiation (which varies according to season and latitude), ocean temperature, and land mass, in addition to many other factors. Since our focus is on the later Pleistocene of mid-latitude Western Europe, we are most concerned with the effects that the major climatic shifts that occurred there had on the frequency of natural fires that European Neandertals could have harvested.

In his review, Sorensen (2017a) presents several different kinds of data to argue that lightning-caused fires did not decrease significantly during the colder periods of the Pleistocene, or at least not to the point that fire would be difficult to obtain. One of his arguments focuses on the frequency of lightning strikes and the probability that any given strike will result in a fire. His basic conclusion is that there were only “modest differences in fire ignition frequencies between climatic periods,” [emphasis added], and, furthermore, that there was a potential “for there to be more fires during a cooler period despite reduced lightning strike frequencies if favourable vegetation and climatic conditions increase ignition rates” (Sorensen 2017a:119–120). He then argues that within the daily hominin foraging range, several natural fires would be present in any given year, even during glacial periods. For him, in other words, an inability of hominins to start fires does not explain the observed variation in fire use simply because fires were always present in the landscape in spite of the marked climate change between glacial and interglacial cycles.

How modest is this association between lightning-caused fires and climate? Again, this is a subject that has been extensively studied, and the consensus answer—discussed even in articles cited by Sorensen—is that it is anything but modest. What is known is that the frequency of natural fires is dependent on two variables: cloud-to-ground lightning frequency and the nature of the vegetation cover (as a source of fuel). In turn, both of these are dependent primarily on temperature and humidity.

Lightning frequency is clearly related to latitude. For example, Table 1 presents data on lightning strike frequencies, and it is clear that lightning strikes are most frequent at the equator and decline steadily with increasing latitude. Not only are more northern areas cooler, they are also drier due to the fact that cooler air cannot carry as much moisture. This is also why lightning strikes are more frequent in the summer and almost nonexistent in the winter (Fig. 1).
Table 1

Range of frequencies of lightning strikes per year for several major global regions


Lightning strikes/km2/year

Notes on relation between lightning and natural fire frequencies

Equatorial Africa


Congo lightning frequency (> 100) is the highest in the world and vegetation is dense, but has very high moisture content



Highest lightning frequency in North America

African Sahel


Low plant biomass and very dry



Low plant biomass and very dry

SW France


Variable plant biomass and wetter

Eurasian Steppe


Low plant biomass and very dry

Alaskan boreal forest


High plant biomass and dry

Northern Alaskan tundra

> 0.1

This equals about 1 strike per km2 every 10 years and each strike will rarely result in a fire (see Wein 1976).

Data from National Oceanic and Atmospheric Administration at

Fig. 1

Correlation between lightning frequency, lightning-caused fire frequency, and season. There is a clear correlation between these three things that results in few or no wild fires during cold months and significant numbers of wild fires in warm months (data from Alaskan Fire Service [AFS] and figure adapted from Farukh and Hayasaka 2012. Fig. 7)

While lightning frequency is a major factor affecting fire frequency, the latter is also dependent on the density and moisture content of aboveground vegetation; both of these are, again, related to climate. Vegetation density, or plant biomass, can be expressed as net primary productivity (NPP), which is mainly a product of the amount of annual sunlight providing the energy for photosynthesis and the overall amount and seasonality of precipitation (Begon et al. 2006). Colder and drier climates have much lower NPP and thus reduced aboveground plant biomass, whereas warmer and more humid areas have more. So, tropical forests have the highest plant biomass and, aside from extreme desert and ice-covered areas, tundra has the lowest. Higher plant biomass does not always result in more lightning-caused fires, however, because vegetation with higher moisture content is less likely to ignite and less likely to fuel the spread of fires. The moisture content of vegetation is largely a product of precipitation and, therefore, climate. Tropical forests have high plant biomass but the vegetation has higher moisture, which, again, leads to fewer natural fires. At the opposite extreme, tundra vegetation tends to be quite dry and flammable, but the frequency of natural fires is greatly reduced because of very low lightning frequencies and a severely limited plant biomass. Areas in the middle latitudes, including steppe grasslands and boreal or deciduous forests, are intermediate in having relatively high biomass, drier vegetation, and moderate frequencies of lightning. It is no surprise, then, that these regions have the highest frequency of lightning-caused fires.

Again, these are all facts that have emerged from studies by climatologists, ecologists, and atmospheric scientists, and there appears to be broad—if not universal—consensus on the relationship between lightning and the frequency of natural fires (see, just a few examples, Archibald et al. 2013, esp. Fig. 2; Ballard et al. 2017; Blarquez et al. 2015; Daniau et al. 2010; Gill et al. 2009; Mooney et al. 2011; van der Kaars et al. 2000, esp. Fig. 3; Montoya et al. 2011). Similarly, recent data (Veraverbeke et al. 2017) and modelling (Romps et al. 2014) also suggest that lightning frequency and lightning-caused wildfires are associated with current global warming. Furthermore, we know that the major cold (glacial) and warm (interglacial) oscillations (and the shorter-term variations within these) that occurred in the Pleistocene clearly resulted in changes in temperature, humidity, and plant biomass and moisture: this has been abundantly documented in the paleontological, palynological, and geological record of Western Europe (e.g., de Beaulieu and Reille 1989; Discamps 2014; Discamps et al. 2011; Goni et al. 2008; Genty 2008; Guiot et al. 1993; Laville 1975; Reille and De Beaulieu 1990; Tzedakis 1994; Woillard 1978). Given what we know about how these variables affect lightning-caused fires today, we can expect that during colder periods in the Pleistocene there was a decrease in both lightning frequencies and the type and quantity of vegetation that would favor the ignition and spread of fire.
Fig. 2

Lightning-caused fires in Alaska in 2002. These fires are restricted in distribution to the Boreal forest and to the summer months. Note that there were no fires in the arctic tundra zone to the north and west of the Brooks Range, which spans the southern edge of the Boreal-tundra zone. Map adapted from Google Earth and data on fires from

Fig. 3

Thin section scans from Pech de l’Azé IV illustrating the difference in the sediments in relation to burned materials. Slides measure 50 × 75 mm. a Thin section from Layer 8 showing, in the upper part, bedded burned bones (arrows) in a dark sandy matrix enriched in fine char. Another layer of bones can be seen in the bottom of the slide. This view is typical for Layer 8, which locally contains ashes and intact hearths. b Thin section from Layer 4C showing two isolated fragments of burned bone (arrows). Note the geogenic nature of the yellow-brown matrix, which contains neither char nor charcoal and is typical for most of the deposits above Layer 5. Out of a total of 27 thin sections from 18 micromorphological samples from Layer 4, this is the only one to show any evidence of burning, thus indicating a dramatic decrease, though not total absence, of burning

In arguing the contrary, Sorensen (2017a) relies on data from Daniau et al.’s (2009) analysis of microcharcoal accumulations in marine cores taken in the Bay of Biscay. This microcharcoal component is mainly the result of terrestrial fires occurring across Western Europe, although the nature of their accumulation is a product of a number of different variables, including changes in ocean currents. First of all, it should be noted that while Sorensen makes use of their data, these authors directly contradict his argument by stating that the “fire regime of western France closely follows Dansgaard-Oeschger climatic variability … namely low fire regime associated with open vegetation during stadials including Heinrich events, and high fire regime associated with open forest during interstadials” (Daniau et al. 2009: 385). Furthermore, if the point is to have a direct proxy for wild fire frequency in the region of Roc de Marsal and Pech IV, then one has to question the relevance of microcharcoal samples that represent a mixture of sediments from the drainages of many river systems (including the Loire–Brittany and the Adour–Garonne river basins in France, as well as those from northern Iberia and the Pyrenees) and off-shore Atlantic currents that collectively pass through a wide range of climatic regions and ecosystems. This sampling problem is not limited to this particular dataset. Verardo and Ruddiman (1996), for example, found that microcharcoal records from North Atlantic cores spanning the last 200,000 years were generally decoupled from climatic conditions and did not accurately reflect terrestrial burning, which for them was unquestionably climate driven (see also Daniau et al. 2013). Taken together, these observations make it questionable whether marine microcharcoal data can be used to estimate fire frequencies in a particular area, such as southwest France, and especially areas further inland.

Likewise, Sorensen’s attempts to model the frequency of lightning strikes and the frequency of fires in the Dordogne valley of France (Sorensen 2017a:119 and his Supplementary Table 1), where Roc de Marsal and Pech IV are located, is not compelling given that his estimates run contrary to virtually all other paleoclimatological data from this area. Admittedly, we are not climatologists, but the lack of agreement between his model and other sources suggest that some of his underlying assumptions are wrong. As stated above, we do know that southwest France during MIS 4 or 3 was predominantly steppe, steppe-tundra, or tundra. Therefore, to estimate the frequency of lightning-caused fire, it is perhaps more appropriate to use data from comparable biomes such as the arctic tundra and shrubland regions of the Northern Slopes and Coastal Plain regions of Alaska, in spite of their higher latitude. As seen in our Fig. 2 (see also Farukh and Hayasaka 2012: Fig. 3), both lightning and natural fire frequencies in this region are so rare as to be almost nonexistent.

Taken altogether, it is almost certainly the case that lightning-caused fires in southwest France were more common during interglacial periods compared to glacial times, and thus the arguments presented by Sorensen are not a refutation of the null hypothesis. One point we do agree with him on, however, is that the frequency of natural fires would not necessarily have dropped to zero for the entire duration of a glacial cycle. First, climates fluctuated throughout the Pleistocene, and it is well known that there were warm spells during glacial periods and cold spells during interglacial ones. Given the imprecision of our dating and the coarseness of the local environmental indicators, such small-scale changes are virtually invisible in the depositional record of our sites. More importantly, and as we have previously reported (Dibble et al. 2017; Sandgathe et al. 2011), our data show that some burning occurred at the sites we studied even during occupations that took place during conditions that were generally cold. Our conclusion was that the frequency or intensity of burning decreased significantly during these times, not that it was completely absent. We will return to this point in the next section.

How Well Does the Evidence Hold Up at Other Sites?

There are two different kinds of arguments that Sorensen (2017a) uses to attempt to show that the evidence from other sites does not support the patterns we observed at Roc de Marsal and Pech IV. One is that some occupations that took place during warm periods do not show higher frequencies of fire use, and the other, based primarily on evidence from the Middle Paleolithic site of Combe Grenal, that fires do occur frequently during cold-climate occupations.

The first of these arguments stems, in part, from an erroneous characterization of our conclusion as to whether or not Neandertals were obligate fire users. He states (2017a:115), for example, that we “cite the sporadic nature of fire use at European MP sites as evidence for Neandertals being obligate fire users”, and he repeats this in other places in the article. He is entirely mistaken, however, in attributing to us this concept of Neandertals as obligate fire users. To the contrary, one of the main conclusions of our work on the late Neandertals of southwest France (Dibble et al. 2017; Sandgathe et al. 2011) is that they were not obligate fire users, based on the very fact that they significantly decreased the use of fire during colder periods. For us, this suggests that Neandertals did not necessarily rely on fire to survive. Thus, some of his arguments based on the assumption that they were obligate fire users are not really relevant to our hypothesis.

One of these arguments, for example, is that if Neandertals were obligate fire users, then they should always be using it when it was available: “if anthropogenic fire signals weaken when it is colder, then they should become stronger when it is warmer. Yet, this is not always the case, as attested to by the near absence of observed fire traces at any number of Last Interglacial (MISS 5e) and Early Glacial sites (MIS 5d-a)” (Sorensen 2017a:121). Our position, which, again, does not assume obligate fire use, is that climate is best understood as a constraint rather than a driver. In cold climates, naturally occurring fire was less available to harvest, thus explaining the decrease in overall fire use. In warm climates, when more naturally occurring fire was available, we expect that other factors, such as site use and mobility, influenced the frequency of fire use. In other words, in our view, anthropogenic fire signals should weaken when it is colder, and become on average higher, though with variable frequency, when it is warmer. Just because these Neandertals appear to have used fire more frequently when it was available, it does not logically follow that they always used it when it was available, and so the fact that some interglacial, and even Holocene sites do not show evidence of burning, is not a direct refutation of our argument. A more appropriate test would be to see if, in multiple sites, there is a general trend toward low occurrence of fire during colder climates and generally higher occurrence in fire use during warmer climates, which would support our hypothesis or conversely, that the pattern is the opposite of what we proposed or that there is no pattern at all. There will always be some degree of variability, of course, in part because of the small-scale climactic variation discussed in the previous section and in part because of a myriad of other factors that may have influenced fire use at particular sites.

It also needs to be made clear that our arguments are not based on presence or absence of fire, but rather relative frequency, just as the words “all” and “nothing” are not the same as “more” and “less”. For example, the two dating papers (Bowman and Sieveking 1983; Vogel and Waterbolk 1967) cited by Sorensen (2017a:117) that analyzed heated flints from layers associated with cold periods at Combe Grenal does not provide any information on the frequency in burning, and, thus, does not refute our thesis. As mentioned in the previous section, we do have small percentages of burned flint and micromorphological evidence for fire (Fig. 3) in almost all levels associated with MIS 4 and 3 at Roc de Marsal and Pech IV, and there is evidence for limited fire use from other sites associated with these climatic periods in southwest France as well, such as La Ferrassie, Chez Pinaud Jonzac (Niven et al. 2012; Richter et al. 2013), La Quina (Bierwirth 1996; Debénath et al. 1999), Pech de l’Azé I (Soressi et al. 2002), and Abri Peyrony (Soressi et al. 2013).

Besides furnishing the few burned flints used for early dating attempts, the site of Combe Grenal, located only a few kilometers from Pech IV and Roc de Marsal, provides much of the archaeological data that Sorensen uses to show a lack of climate-related patterning at other sites. Combe Grenal is a deeply-stratified, multi-layered Middle Paleolithic site, excavated by Bordes in the 1950s and 1960s (Bordes 1972). Sorensen (2017a:121) states: “In all, 127 small and 19 large hearths (foyers in French) were documented by Bordes in his excavation notebooks (Binford 2007) … with three-quarters of these in ‘cold’ layers.” Superficially, this would appear to contradict the patterns we showed for Pech IV and Roc de Marsal. There are, however, a number of problems with using these data, as published by Binford, to test our hypothesis. First, it is important to note that the French word “foyer” has a multitude of meanings, and even in an archaeological context the word may have nothing to do with fire. Second, and more importantly, our experience with Bordes’ field notebooks from both his Combe Grenal (Dibble et al. 2009; McPherron et al. 2012) and Pech IV excavations (McPherron and Dibble 1999) and the experience of one of us (HLD) excavating with Bordes in the 1970s at Pech IV (Dibble et al. 2018) have shown that Bordes’ field notebooks are fraught with errors. In the first place, the excavators were virtually all students who, at the time they were working with Bordes, had widely different levels of experience and training in archaeology. Furthermore, there was little or no on-site supervision or adherence to specific protocols—the nature and content of any notes they took was left entirely up to them (McPherron and Dibble 1999). Thus, relying on their notebooks as a primary data source for what they saw while they were digging is seriously problematic, and there is little or no way to differentiate solid observations from uninformed opinions or mere guesses. The fact is, based on our experience working on the Combe Grenal collections (Dibble et al. 2009) and on Bordes’ field notebooks, there are many reasons to be skeptical of much of the documentation that remains for this site.

Moreover, there are numerous examples in Paleolithic archaeology where features that superficially resembled hearths turned out to be something unrelated to fire (Goldberg et al. 2018; Goldberg et al. 2001; Weiner et al. 1998)—manganese in the case of Fontéchevade (Chase 2009), diagenesis in the case of Zhoukoudian layers 4 and 10 (Goldberg et al. 2001; Weiner et al. 1998), and organic residues in the case of Schöningen (Stahlschmidt et al. 2015). At La Cotte de St. Brelade, Callow et al. (1986), reported that “[p]ossible hearths were initially noted in all layers … [b]ut detailed analysis indicates that not one of these can be confirmed as a true hearth in situ. All may be explained as geomorphologically or geochemically produced features, or products of interaction between the excavation process and site topography.” Even what Sorensen (2017a:121) cites as “fire-reddened bedrock” under layer 55 at Combe Grenal is problematic. While rubefication of underlying substrates (sediments or bedrock) can be the result of exposure to heat (Aldeias 2017; Aldeias et al. 2016), this is not the only possible cause. In fact, similar features on the bedrock floor at Pech IV, what we initially assumed to be hearth locations, turned out to be the result of the diagenesis and alteration of organic matter that had nothing to do with heating by fire (Goldberg et al. 2018:25). Other examples include red banded staining on limestone blocks at Roc de Marsal (Sandgathe et al. 2004:175), bright red diagenetic alteration of Middle Paleolithic deposits at Theopetra Cave, Greece (Karkanas 2001), and the wide-spread red banding at Grotte XVI, France (Karkanas 2001; Karkanas et al. 2002; Karkanas personal communication). For all of these reasons, we do not think entries in Bordes’ notebooks, which reflected superficial observations made at the time of excavation, are a good source of data to test our hypothesis.

We recently did our own study of fire use through time at Combe Grenal using the same methodology we have applied elsewhere, namely calculating the percentage of heated lithics (Dibble et al. 2017). The results (Fig. 4) show a pattern similar to what we had already found at the nearby sites of Roc de Marsal and Pech IV: there are substantial periods in the sequence exhibiting low percentages of heated lithics, which is evidence for less frequent fires, and, with the exception of Layer 20, the greater evidence for fire is limited to MIS 5 or earlier deposits that are associated with milder climates. Note that the Combe Grenal sequence is undated, so our climate reconstructions and chronology are based on the work of Morin et al. (2014) and Discamps (2011; Discamps and Royer 2017), as well as the previously published palynological (Paquereau 1974), paleontological (Bordes et al. 1966; Chase 1986; Guadelli and Laville 1988; Bouchud 1966; Delpech and Prat 1995), and geological (Laville 1975) evidence.
Fig. 4

Percentages of burned flints per level at Combe Grenal compared to both early climatic interpretations (based on pollen, fauna, and geological data) and more recent reassessments of Bordes’ faunal collection. (Note: for each layer studied all of the lithic material was examined, but counts are based on platform-bearing flakes only to control for increased fragmentation resulting directly from heating)

Again, however, it should be emphasized that there are other issues surrounding the Combe Grenal evidence. Recently, Discamps and Faivre (2017) presented data from two newly re-excavated levels at Combe Grenal that suggests collection bias in the Bordes faunal collection, serious enough that they call into question previous efforts to use the faunal sequence from Combe Grenal to reconstruct climate or hunting behaviors. Similarly, there are curatorial problems for the lithics collections that have resulted in the mixing of materials from different levels (Dibble et al. 2009). The recently renewed excavations (Discamps and Faivre 2017) at this site will undoubtedly clarify the sequence, its chronology, the nature of the lithic and faunal assemblages, and the frequency of evidence of fire use.

Nonetheless, the example from Combe Grenal remains important. First, it is a deep sequence with a detailed record of the last glaciation. Second, it provides another example where different climatic periods are represented at the same site (allowing control of other variables like site situation). Third, these climatic periods vary during the course of the sequence, thereby allowing the association of climate and use of fire to be tested against major changes in site morphology (see below). Thus, rather than refuting the relationship between climate and frequency of fire use, the available Combe Grenal evidence, in spite of its limitations, supports our earlier findings. Together with other sites having comparable data (see Fig. 5), with very few exceptions, climate remains a strong predictor for the use of fire at all these sites.
Fig. 5

Association between burned flint percentages (by level) at several Middle Paleolithic sites in southwest France and major climatic periods (pollen record adapted from Reille and De Beaulieu, 1990)

It is also important to point out that not all types of data on natural fire frequencies are suitable proxies for hominin use of fire. For example, Sorensen (2017a: 121) compared microcharcoal concentrations from marine cores directly to percentages of burned lithics from an archaeological site as a test for the efficacy of the latter to serve as a measure of Neandertal fire use frequency. These are independent datasets with very different formation processes, very different measures, and very different scales and contexts. The microcharcoal record reflects the deposition of material from an exceedingly large source area, while the burned flint data reflects completely different types of accumulation processes that occur under very local conditions specific to individual strata, hearth locations, and human behavior at the site scale. Similarly, there is no reason to expect that percentages of reindeer from the different strata in Roc de Marsal and Pech IV would strongly correlate with the percentages of burned objects (Sorensen 2017a: 122). Even taken on its own, the number of reindeer in a landscape is not an absolute reflection of any specific climatic parameter (e.g., a 30% increase in reindeer does not mean that there was a 30% decrease in temperature). Likewise, the number of reindeer remains recovered from an assemblage reflects not just their presence in the landscape, but a multitude of other factors as well, including human hunting patterns. So, even if we are right—that Neandertal use of fire is related to climate—we would never suggest that for each degree of temperature increase or decrease fire use would increase or decrease by a set amount. Likewise, in our Fig. 5, there is no basis for correlating every percentage change in arboreal pollen with percentage changes in burned lithics. The structure of our argument is much different: a prevalence of reindeer and non-arboreal pollen indicate cold conditions and these assemblages have less evidence of fire use. Conversely, assemblages clearly dominated by red deer, roe deer, and wild pig, for example, have more frequent evidence of fire use.

Our argument obviously relies heavily on our ability to correlate lithic assemblages and burned flint frequencies with climatic conditions. We do not, however (contra Sorensen 2017a: 121–122), correlate stratigraphic sequences with climatic regimes simply on the basis of absolute chronology. Based on a number of recent numeric dates obtained through a variety of techniques (Guérin et al. 2012; Jankowski 2018), we are able to demonstrate the overlap in the sequences of Roc de Marsal and Pech IV and tentatively assign them to various OIS stages. However, we used other data, such as macrofauna (Hodgkins 2018; Hodgkins et al. 2016; Niven 2013, 2018; Castel et al. 2016), microfauna (Marquay as reported in Sandgathe et al. 2008), phytoliths (Cabanes as reported in Sandgathe et al. 2008), and sedimentology (Sandgathe et al. 2008; Goldberg et al. 2018) to reach conclusions about the climate at the time a layer was being deposited and the occupations took place. Not only at Roc de Marsal and Pech IV, but also at other sites, for example, La Ferrassie, La Quina, Chez Pinaud Jonzac, and Abri Peyrony, there is particularly good faunal preservation resulting in significant assemblages for paleontological analysis and paleoclimatic reconstruction, corroborated by chronometric dates and geological evidence. Our arguments about climate during the occupations are, therefore, based strongly on data representing conditions present in the immediate surrounding region at the time the individual layers were formed and occupied. Thus, regardless of the dates for these or any other sites, the layers with significant evidence for fire are associated with other evidence reflecting warm climatic conditions and the layers with very little evidence for fire were formed during cold climates.

Finally, while this issue was generally addressed in our previous publications (Aldeias et al. 2012; Goldberg et al. 2012; Sandgathe et al. 2011, Goldberg et al. 2018), Sorensen (2017a: 124) suggests that we might have missed potential evidence for fire in the upper layers of the Roc de Marsal and Pech IV because of “… the localised nature of [micromorphology] samples and the propensity for researchers to sample locations where (fire) features are plainly visible.” While we are unaware of such a propensity, we can clarify that during our excavations we took 148 and 93 micromorphological samples, respectively, from the main excavation areas of these sites (Table 2 and Figs. 3, 6, and 7). The goals of the micromorphological sampling and analyses were comprehensive and, in large part, geared to understanding site history in all its aspects, and certainly layers without other evidence of burning were sampled extensively (see also Fig. 3).
Table 2

Number of micromorphological samples by layer at Pech de l’Azé IV and Roc de Marsal

Note that at Roc de Marsal some larger samples spanned multiple levels so the total here will be greater than the actual number of samples taken for the site

Fig. 6

Distribution of micromorphological samples taken during the 2000–2003 excavations at Pech IV projected onto a drawing of the west section

Fig. 7

Vertical and horizontal distributions of micromorphological samples taken during the 2004–2010 excavations at Roc de Marsal projected onto a schematic drawing of the west section

Are There Other Explanations for the Relationship Between Climate and Fire Use at These Particular Sites?

So far, we have covered two major classes of arguments that have been put forward by Sorensen (2017a) to refute the patterns we presented, and in both cases these arguments fall short: lightning-caused fires do vary with climate and data from other nearby sites do show similar patterns. Nonetheless, these patterns require explanation, if for no other reason than they cast doubt on the potential role that fire played in adaptations to the varying climatic conditions of the last glacial period.

While our hypothesis concerns the ability—that is, the technological know-how—to make fire, many other scenarios could be proposed to explain the patterns we see in these sites. In our original paper (Sandgathe et al. 2011), we reviewed several: taphonomic processes that may have removed evidence for fire use, object density, site use, excavator bias, etc., and rather than repeat those arguments here, the reader is encouraged to refer to that publication. Since then, others have offered additional alternative explanations, and it is worthwhile to review several of these here.

Changes in Site Structure, Such as Roof Collapse, that May Have Affected How the Inhabitants Used the Site (e.g., Sorensen 2017a: 123)

First, in order to advance such an explanation to explain variation in fire frequency, some additional explanation is required as to how such changes would actually affect fire use at a particular site. Second, based on quantities of lithics and faunal remains at both Pech IV and Roc de Marsal, there is no indication that intensity of occupation (as evidenced by the density of objects; see Fig. 7 in Sandgathe et al. 2011) was affected by changes in cave morphology. Furthermore, while major changes to the caves do occur, there is no indication that they had any effect on fire frequencies. At Roc de Marsal, the only major change occurred as a single major roof fall that happened after the final Middle Paleolithic occupations. At Pech IV, the first of the major collapses are visible only the western section of the site and occurred during the deposition of Layer 6, which includes significant evidence for fire use; later collapses occurred during and after the final Middle Paleolithic occupations. Finally, the Combe Grenal sequence, with its repeated cycles of climatic change, runs counter to the argument that changes in the cave’s morphology affected fire use—it would be incredulous to argue that each onset of cold conditions was accompanied by structural modifications that inhibited fire use and during each warmer period other changes occurred that encouraged it.

Changes in Mobility or Duration of Site Occupations Across the Landscape in Response to Changing Environmental Conditions (e.g., Sorensen 2017a:122)

An explanation such as this, especially to the extent that it is a result of climatic variation, could explain the coincidental patterning at specific sites. As discussed above and in our original publication, however, intensity of occupation does not change through the sequences at Pech IV and Roc de Marsal. Every indication (artifact density, faunal remain density, faunal element representation, butchering patterns) is that both sites were primarily habitation sites during both cold and warm periods (Sandgathe et al. 2011). Moreover, it remains necessary to document the widespread occurrence of fire at other contemporaneous sites in the region. That is to say, if Neandertals were occupying Pech IV and Roc de Marsal but not using fire at them, then where were they using fire?

Changes in the Availability of Fuels Necessary for Fire (Sorensen 2017a: 124)

As pointed out by Henry (2017), fuel availability is likely not just a question of whether adequate fuel was present or absent, but rather as a balance between the benefits of fire use vs. the costs of obtaining fuel. This is another explanation that could be influenced by climatic changes, and could, therefore, be applicable to multiple sites in a region. If this explanation holds, it brings us back to the question that we raised earlier (see also Dibble et al. 2018): if the benefits of fire did not outweigh its cost, then fire cannot be seen as a requisite component of hominin adaptations in Europe during the last glacial.

Coupled with the Availability of Fuel Is the Availability of Suitable Tinder Necessary for Starting Fires with Traditional Methods

Strike-a-light methods produce a spark, but suitable tinder is needed to catch and propagate it: without suitable tinder, starting a fire with such methods is extremely difficult. Ethnohistorically, people commonly relied heavily on certain species of fungi as tinder (e.g., Fomes fomentarius and Haploporus odorus, which grow on trees and mainly deciduous species) (Schmidt 2006; Blanchette 1997). While serviceable alternatives may be available in the cold, open biomes of Pleistocene Europe, climate may have played a role in the availability of suitable tinder and, thus, the ability to start a fire.

Alternatives to Cooking?

There is also a possibility that colder periods were associated with a different way of processing meat, for instance, allowing it to putrefy and ferment, thereby increasing its nutritional values without cooking (Speth 2017). In itself, this may have decreased the need for fire for this purpose, though it still implies that using fire for other functions seen in modern hunter gathers, like warmth or light, for instance, was not an essential part of European Neandertal adaptation to colder climates.

The Use of Small Fires that Would Not Leave Recognizable Traces

Another possible alternative that we will review in more depth is that at certain times hominins built ephemeral fires of limited size or duration: “a reduction in fire signal during cold climatic episodes does not necessarily indicate a climatically mediated reduction in fire use; instead shifts in fauna … and vegetation … triggered cultural responses that resulted in a fundamental change in how and when fire was used, the practice likely becoming more ephemeral but no less important or regular in its use” (Sorensen 2017a: 125). In this scenario, a reduction in the percentage of burned objects would not necessarily reflect a lower frequency of fire during occupation, but only that the fires were not of sufficient temperature or duration to thermally alter underlying objects. This alternative explanation can be explored with data from Roc de Marsal and Pech IV.

It is first necessary to define an ephemeral fire. For the more common and important uses of fire—for example, cooking, warmth, light, or protection from predators—there would necessarily be a minimum size and duration of a fire required to perform the function—notwithstanding extremely specialized functions such as lighting a cigarette (Mallol et al. 2007) or, as observed by one of us (PG), Bedouin tea making. Here we will assume circular fires of either 50 or 25 cm diameter that attained a temperature of 600 °C for 2 h. Controlled heating experiments by Aldeias et al. (2016) show that the transfer of heat into underlying sediments proceeds downward in a spherical cap, or bowl-shape, with the rate of heat transfer and depth of penetration primarily a function of both fire temperature and duration; it is also affected to a lesser degree by type of sediment, moisture content, and presence of organic matter. Based on these results, fires of this size, temperature, and duration can heat underlying sediments to approximately 300 °C to a depth of 4 cm, with higher temperatures reached at shallower depths. At temperatures of around 300 °C, both lithics and bones are visibly altered (e.g., Julig et al. 1999; Stiner et al. 1995). Parenthetically, although Sorensen (2017a: 123) suggests that different fuel types would affect the rate of heat transfer, our understanding of the physics of heat transfer is that it relates almost exclusively to temperature differential and conductivity, not to the source of heat.

Using these assumptions, we can then estimate the number of subsurface objects that would be thermally altered by a fire of a specific size, temperature, and duration. The volume of sediments heated to the maximum (within the bowl-shaped spherical cap directly underneath a fire) would be approximately four liters for a 50-cm fire and a little over 1 l for a 25-cm fire. The number of heated objects, assuming that they are distributed randomly (which is generally the case at both Roc de Marsal and Pech IV), would then be a function of object density. From there, by knowing the actual number of burned objects in a layer, it is possible to estimate the minimum number of fires of this size that had occurred over the course of occupations associated with that layer. Table 3 presents data from Roc de Marsal and Pech IV. For example, Roc de Marsal Layer 9 had 12.2 objects per liter of sediments, which means there would be, on average, 48.3 objects inside the volume of sediments heated below a 50-cm fire and 12.4 objects for a 25-cm fire. The total of 3831 burned objects from Layer 9 provides an estimate of a minimum of 79 fires for the whole of the Layer 9 occupations (but see note below Table 3). For Roc de Marsal Layer 4, there would have been fewer than two fires of 50 cm size, or just over six 25 cm-size fires, over millennia of occupations.
Table 3

Estimation of the number of fires of different sizes (all attaining a temperature of 600 °C and a 4-cm depth of heating) that would account for the number of burned objects (lithics and bones) in several example layers at Roc de Marsal (RDM) and Pech IV. For the 50 and 25 cm diameter fires, the volumes of the spherical caps would be 3.96 and 1.015 l, respectively

Site and layer

Total volume of sediments (liters)

Total lithics recovered

Total lithics/liter

N burned lithics

50 cm fire heating to a depth of 4 cm

25 cm fire heating to a depth of 4 cm

N per spherical cap (3.96 L)

Estimated number of fires

N per spherical cap (1.02 L)

Estimated number of fires



















Pech IV 8









Pech IV 4c









Pech IV 3a









Note: Two important factors need to be considered in interpreting the values in the column “Estimated number of fires in level.” The first is that they are based on all burned lithics in the layer, but it is known that significant heating increases fragmentation rates—single flakes are often fragmented into one or more pieces after burning—which means the numbers will be inflated somewhat. However, this issue will apply to all levels regardless of overall fire frequencies. The second factor is one that realistically affects only levels with high frequencies of fires: since multiple fires can be built in more or less the same location (clearly the case for Layer 9 at Roc de Marsal and Layer 8 at Pech IV) many burned objects will have been heated more than once. Thus, for these layers the numbers will underestimate actual fire frequency

Of course, these estimates depend on certain assumptions, but it does help to illustrate an important point: in these sites, where artifact densities are very high, even relatively small or ephemeral fires (in terms of size, temperature, and duration) will thermally alter objects and having more such fires will inevitably result in a greater accumulation of burned objects. There is, in addition, a limit to how small a fire can reasonably be—after a point they will not function for normal uses. A fire of 25 cm in diameter is already very small, but given our high artifact densities, several such fires would still translate into a much larger number of burned objects than what we recovered in the cold-climate layers.

Did Neandertals Have Fire-Making Equipment?

It is one thing to argue that various characteristics of Paleolithic fires would or would not be consistent with an ability to start fires, but having evidence of fire-making equipment would be, so to speak, the smoking gun. However, the identification of fire-making equipment in the archaeological record is not straightforward. Beyond issues of preservation and site formation, materials that have the potential to be useful for igniting fires should also appear in contexts associated with other evidence for fire.

From what we know of traditional means of starting fires (see Sorensen et al. 2014), the equipment involved in friction-based methods would not be expected to survive except in exceptional circumstances. However, minerals such as pyrite or marcasite can produce a spark, and Sorensen (2017a: 125) cites examples of such minerals found at various European sites. It is important to bear in mind that most Paleolithic sites contain a certain number of allochthonous rocks and minerals, some entering the sites naturally (including being released from the bedrock—e.g., Aldeias et al. 2014) and some through human action—often it is difficult to determine which. Thus, depending on the context, the presence of, say, a nodule of pyrite is not, in itself, evidence that it was used for any particular function. To make a stronger case that these minerals were used to start fires, it would be necessary to recover them in sufficient quantity and in association with other evidence of fire—we would expect that the regular use of them for starting fires should result in a notable presence of small and large fragments, especially in a context of hearth-rich deposits. As it is, only one piece of marcasite was recovered from Roc de Marsal (from a layer with almost no evidence for fire), and no such minerals were recovered or have yet been identified from Pech IV (in either Bordes’ or our excavations), Combe-Capelle Bas, La Ferrassie, or Abri Peyrony.

More indirect, though still potentially relevant, evidence for fire making may be manganese dioxide (MnO2), which Heyes et al. (2016) suggest could have served as a fire accelerant—in powdered form, manganese dioxide can reduce the temperatures required for ignition. Again, its presence in a site is not, in itself, a demonstration that it was used for this purpose, in spite of its potential benefits. But, one way to test this would be to examine its presence in specific contexts: is its presence in sites associated with the other evidence for the use of fire? At Roc de Marsal, not a single piece of manganese of any size was recovered from any layers (either among the provenienced objects or from the 6 mm wet-screened fraction), and there are no records of any found during the original excavation. Notable quantities of manganese (26 pieces in total, 4 of which have no stratigraphic information) were recovered from Pech IV during both Bordes’ and our own excavations, and some of these have surface modifications (Soressi and d'Errico 2007; d’Errico and Soressi 2018) that suggest they were serving some purpose for the hominins living there. That purpose, however, does not appear to have been related to fire. In fact, when the frequency of manganese is plotted by layer (Fig. 8), there appears to be an inverse relationship between the occurrence of this mineral and evidence for fire, either in terms of the quantity of burned flint (shown in the figure) or other evidence such as ash, charcoal, and burned bone. Although it is not known if the composition of the Pech IV manganese is the accelerant variant (Heyes et al. 2016), it is clear that there is no association between the presence of this mineral and fire.
Fig. 8

Frequency at Pech IV (by level) of pieces of manganese (provenienced in recovered from coarse (6 mm) fraction of wet-screened sediments) and percentages of platform-bearing lithics that were burned

It has also been proposed (Sorensen 2017b) that the surfaces of bifaces may have been used to produce sparks for igniting fires. While they may have the potential to function in this regard, if they were actually used for this purpose we would, again, expect to see their occurrence more often in contexts that had increased evidence for fire. Figure 9 presents data from several assemblages on the percentage of bifaces and the percentage of burned lithics, and no such association is apparent.
Fig. 9

Percentage of bifaces (complete and partial) and percentages of burned flints, with both axes scaled logarithmically for clarity. Biface percentages computed as Bordes’ (Debénath and Dibble 1994) Biface Index, or the number of bifaces relative to the total of bifaces and retouched flakes. Assemblages include those from Abri Peyrony (unpublished), Combe-Capelle Bas (Dibble and Lenoir 1995), Combe Grenal (data from Bordes’ unpublished type counts), La Ferrassie (unpublished), La Quina (Jelinek 2013), Pech IV (McPherron et al. 2018), and Roc de Marsal (unpublished)

This does not prove that bifaces, or particular minerals in the earlier examples, were not used for this purpose (again, that would be an argument from ignorance), but nonetheless seeing them associated with the presence of fire is a reasonable and valid test implication. These examples illustrate that a demonstration that certain materials or objects were used to start fires depends not just on demonstrating their potential for this purpose, but also a demonstration that they occur in particular contexts in association with other evidence for fire.


The continued and, perhaps, accelerating development of new technologies is one of the hallmarks of the human lineage, and from our current vantage point it would seem that pyrotechnology is certainly among the more important ones. It is rare, though, for a new technology to achieve its full potential immediately, and so it is likely that hominin use of fire also advanced through a number of steps, from casual and opportunistic use of natural fires to the ability to transport, contain, and fuel it to, ultimately, the ability to start it (Sandgathe 2017). The presence of isolated combustion features, or hearths, in many European Middle Paleolithic occupations attests to the ability of the Neandertals there to control fire, and such evidence has naturally led to an assumption that they had the ability to start fires as well. What has called this assumption into question is the recent finding, at least in this part of Western Europe, that the frequency of fire varies, seemingly as a response to climatic change: during warmer periods, when the frequency of natural fires is highest, evidence of fire in Middle Paleolithic sites is relatively common, while during colder periods, when natural fires are less frequent, there is a dramatic reduction in the evidence of fire in occupation sites. One hypothesis to explain this pattern is that these Neandertals lacked the technological knowledge to make fire themselves but were able to gather naturally occurring fire when it was available and transport it into their sites. During colder periods, a general reduction in naturally occurring fires constrained its use.

Based on what is known to affect the occurrence of natural fires today, which are primarily caused by lightning, there is no compelling reason to doubt that colder periods during the Pleistocene of mid-latitude Western Europe led to a reduction in natural fires. While there are other potential explanations for varying frequency of use of fire during the European Middle Paleolithic, they must also account for the association with climate. However, in addressing this question, it is necessary to move beyond simple presence or absence of fire evidence to a more quantitative approach that can provide estimates for frequency of fire use. Likewise, when investigating possible fire-starting technologies, it is necessary to show a clear association between their presence and the presence of fires.

In this review, we have attempted to address many arguments that have been raised in reaction to earlier presentations regarding the patterns in fire use observed in Middle Paleolithic sites we have excavated and studied. Our conclusion here is that the null hypothesis—that these populations lacked the ability to make fire—still stands as the best explanation for the patterns observed there.

Beyond this specific conclusion, there are three additional points to be emphasized. First, a demonstration of fire making by some Neandertals at an earlier date or in a different region does not negate this hypothesis. Again, what we are investigating is whether the Neandertals that lived in southwest France during the Late Pleistocene could start fires and whether they were obligate fire users. Since human technologies are independent of biology, the presence or absence of a particular technology in one population does not necessarily mean that the same is true for others. Indeed, if it were demonstrated that some Neandertal groups could start fires, or were fully reliant on pyrotechnology, then this would be a significant step forward in understanding technological variability among Neandertals. The second point to be emphasized is that a lack of knowledge concerning a specific technology has no implications for their cognitive abilities or their taxonomic relationship to modern humans—even the recent history of the world over the past several millennia is replete with instances of some people lacking particular technologies that others had already developed. Finally, while an inability to make fire can explain the apparent link between climate and the variation in fire frequency in Neandertal sites in southwest France, it does not inform us as to the nature of their adaptations. This is why the finding of reduced use of fire during colder periods remains of considerable importance, since it has long been assumed that fire was one of the primary means that allowed hominin expansion into Pleistocene Europe and their continued presence even during extreme climatic shifts. This, then, is a fundamental question we should be addressing, since it forces us to begin investigating other possible means that allowed Neandertals, and perhaps earlier hominins as well, to adapt to those situations.



Chez Pinaud Jonzac was excavated by Jacques-Jaubert and Jean-Jacques Hublin with Marie Soressi and SPM. All are thanked for agreeing to share this portion of the lithic data from Jonzac. Abri Peyrony was excavated by Michel Lenoir and SPM. Tamara Dogandžić supervised the field work and studies the lithics with SPM. Again, all are thanked for agreeing to share this portion of the lithic data. SPM thanks the Max Planck Society and Jean-Jacques Hublin for support of this research. Aylar Abdolahzadeh was responsible for most of the collection of the Combe Grenal burned flint data, and we would like to thank Stéphane Madelaine of the Musée National de Préhistoire for arranging access to this material. Much of the data used here was the result of research projects that were possible through funding provided by the US National Science Foundation, the Leakey Foundation, Service Régional d’Archéologie, the Conseil Général de la Dordogne, Max Planck Institute for Evolutionary Anthropology, National Geographic Society, the University of Pennsylvania Research Foundation, the University of Pennsylvania Museum of Archaeology and Anthropology, and the Marie Curie International Fellowship within the 6th European Community Framework Program.

Compliance with Ethical Standards

Conflict of Interest

On behalf of all authors, the corresponding author states that there are no conflicts of interest.


  1. Aldeias, V. (2017). Experimental approaches to archaeological fire features and their behavioral relevance. Current Anthropology, 58(S16), S191–S205.CrossRefGoogle Scholar
  2. Aldeias, V., Dibble, H. L., Sandgathe, D., Goldberg, P., & McPherron, S. J. (2016). How heat alters underlying deposits and implications for archaeological fire features: a controlled experiment. Journal of Archaeological Science, 67, 64–79.CrossRefGoogle Scholar
  3. Aldeias, V., Goldberg, P., Dibble, H. L., & El-Hajraoui, M. (2014). Deciphering site formation processes through soil micromorphology at Contrebandiers Cave, Morocco. Journal of Human Evolution, 69, 8–30.CrossRefGoogle Scholar
  4. Aldeias, V., Goldberg, P., Sandgathe, D., Berna, F., Dibble, H. L., McPherron, S. P., et al. (2012). Evidence for Neandertal use of fire at Roc de Marsal (France). Journal of Archaeological Science, 39(7), 2414–2423.CrossRefGoogle Scholar
  5. Archibald, S., Lehmann, C. E., Gómez-Dans, J. L., & Bradstock, R. A. (2013). Defining pyromes and global syndromes of fire regimes. Proceedings of the National Academy of Sciences, 110(16), 6442–6447.CrossRefGoogle Scholar
  6. Ballard, J. P., Horn, S. P., & Li, Z.-H. (2017). A 23,000-year microscopic charcoal record from Anderson Pond, Tennessee, USA. Palynology, 41(1–2), 216–229.CrossRefGoogle Scholar
  7. Barbetti, M. (1986). Traces of fire in the archaeological record, before one million years ago? Journal of Human Evolution, 15(8), 771–781.CrossRefGoogle Scholar
  8. Barkai, R., Rosell, J., Blasco, R., & Gopher, A. (2017). Fire for a reason: barbecue at Middle Pleistocene Qesem Cave, Israel. Current Anthropology, 58(S16), S314–S328.CrossRefGoogle Scholar
  9. Begon, M., Townsend, C. R. H., John, L., Colin, R. T., & John, L. H. (2006). Ecology: from individuals to ecosystems (Vol. Sirsi).Google Scholar
  10. Berna, F., Goldberg, P., Horwitz, L. K., Brink, J., Holt, S., Bamford, M., et al. (2012). Microstratigraphic evidence of in situ fire in the Acheulean strata of Wonderwerk Cave, Northern Cape province, South Africa. PNAS, 109(20), 7593–7594.CrossRefGoogle Scholar
  11. Bierwirth, S. L. (1996). Lithic analysis in southwestern France: Middle Paleolithic assemblages from the site of la Quina. Oxford: British Archaeological Reports Ltd..Google Scholar
  12. Binford, L. R. (2007). The diet of early hominins: Some things we need to know before “reading” the menu from the archaeological record. Guts and brains: An integrative approach to the hominin record, 185–222.Google Scholar
  13. Blanchette, R. A. (1997). Haploporus odorus: a sacred fungus in traditional Native American culture of the northern plains. Mycologia, 89(2), 233–240.CrossRefGoogle Scholar
  14. Blarquez, O., Ali, A. A., Girardin, M. P., Grondin, P., Fréchette, B., Bergeron, Y., et al. (2015). Regional paleofire regimes affected by non-uniform climate, vegetation and human drivers. Scientific Reports, 5, 13356.CrossRefGoogle Scholar
  15. Bordes, F. (1972). A Tale of Two Caves. New York: HarperCollins Publishers.Google Scholar
  16. Bordes, F., Laville, H., & Paquereau, M. M. (1966). Observations sur le Pléistocène supérieur du gisement de Combe-Grenal (Dordogne). Actes de la Société Linnéenne de Bordeaux, 103(10), 3–19.Google Scholar
  17. Bouchud, J. (1966). Essai sur le Renne et la climatologie du Paléolithique moyen et supérieur. Imprimerie Magne.Google Scholar
  18. Bowman, S. G. E., & Sieveking, G. D. G. (1983). Thermoluminescence dating of burnt flint from Combe Grenal. Journal of the European Study Group on Physical, Chemical and Mathematical Techniques Applied to Archaeology, 9, 253–268.Google Scholar
  19. Callow, P., Walton, D., Shell, C., Callow, P., & Cornford, J. (1986) The use of fire at La Cotte de St. Brelade. La Cotte de St. Brelade 1961–1978. Excavations by CBM McBurney, 193–195.Google Scholar
  20. Castel, J.-C., Discamps, E., Soulier, M.-C., Sandgathe, D., Dibble, H. L., McPherron, S. J. P., et al. (2016). Neandertal subsistence strategies during the Quina Mousterian at Roc de Marsal (France). Quaternary International, 433(Part B), 140–156.Google Scholar
  21. Chase, P. G. (1986). Relationships between Mousterian lithic and faunal assemblages at Combe Grenal. Current Anthropology, 27(1), 69–71.CrossRefGoogle Scholar
  22. Chase, P. G. (2009). The Cave of Fontéchevade: recent excavations and their paleoanthropological implications. Cambridge: Cambridge University Press.Google Scholar
  23. Courty, M. A., Carbonell, E., Vallverdú Poch, J., & Banerjee, R. (2012). Microstratigraphic and multi-analytical evidence for advanced Neanderthal pyrotechnology at Abric Romani (Capellades, Spain). Quaternary International, 247(1), 294–312.CrossRefGoogle Scholar
  24. Daniau, A.-L., Goñi, M. F. S., & Duprat, J. (2009). Last glacial fire regime variability in western France inferred from microcharcoal preserved in core MD04-2845, Bay of Biscay. Quaternary Research, 71(3), 385–396.CrossRefGoogle Scholar
  25. Daniau, A.-L., Goñi, M. F. S., Martinez, P., Urrego, D. H., Bout-Roumazeilles, V., Desprat, S., et al. (2013). Orbital-scale climate forcing of grassland burning in southern Africa. PNAS, 110(13), 5069–5073.CrossRefGoogle Scholar
  26. Daniau, A.-L., Harrison, S., & Bartlein, P. (2010). Fire regimes during the last glacial. Quaternary Science Reviews, 29(21), 2918–2930.CrossRefGoogle Scholar
  27. de Beaulieu, J.-L., & Reille, M. (1989). The transition from temperate phases to stadials in the long Upper Pleistocene sequence from Les Echets (France). Palaeogeography, Palaeoclimatology, Palaeoecology, 72, 147–159.CrossRefGoogle Scholar
  28. Debénath, A., & Dibble, H. (1994). Handbook of Paleolithic typology Vol 1: The lower and middle Paleolithic of Europe. Philadelphia: The University Museum, University of Pennsylvania. ix + 202 p.Google Scholar
  29. Debénath, A., Jelinek, A., Armand, D., Chase, P., Dibble, H., Mercier, N., Renault-Miskovsky, J., Tillier, A.-M., Valladas, H., & Vandermeersch, B. (1999). Nouvelles Fouilles à La Quina (Charente). Gallia Préhistoire, 40, 29–74.Google Scholar
  30. Delpech, F., & Prat, F. (1995). Nouvelles observations sur les faunes acheuléennes de Combe-Grenal (Domme, Dordogne). Paleo, 7(1), 123–137.Google Scholar
  31. Dibble, H. L., Aldeias, V., Goldberg, P., McPherron, S. P., & Sandgathe, D. (2017). How did Hominins adapt to Ice Age Europe without fire? Current Ahtropology, 58(S16), S278–S287.CrossRefGoogle Scholar
  32. Dibble, H. L., Berna, F., Goldberg, P., McPherron, S., Mentzer, S., Niven, L. et al. (2009a). A preliminary report on Pech de l'Azé IV, layer 8 (Middle Paleolithic, France). PaleoAnthropology, 182–219Google Scholar
  33. Dibble, H. L., McPherron, S. P., Sandgathe, D., Goldberg, P., Turq, A., & Lenoir, M. (2009b). Context, curation, and bias: an evaluation of the middle Paleolithic collections of Combe-Grenal (France). Journal of Archaeological Science, 36(11), 2540–2550.CrossRefGoogle Scholar
  34. Dibble, H., & Lenoir, M. (1995). The Middle Paleolithic Site of Combe-Capelle Bas (France). Philadelphia: University Museum Press, University of Pennsylvania xviii+363 pp.Google Scholar
  35. Dibble, H. L., McPherron, S., Goldberg, P., and Sandgathe, D. (2018) The Middle Paleolithic Site of Pech de l’Azé IV. Monograph: Cave and Karst Systems of the World Series. Springer.Google Scholar
  36. Discamps, E. (2014). Ungulate biomass fluctuations endured by Middle and Early Upper Paleolithic societies (SW France, MIS 5-3): The contributions of modern analogs and cave hyena paleodemography. Quaternary International, 337, 64–79.CrossRefGoogle Scholar
  37. Discamps, E., & Faivre, J.-P. (2017). Substantial biases affecting Combe-Grenal faunal record cast doubts on previous models of Neanderthal subsistence and environmental context. Journal of Archaeological Science, 81, 128–132.CrossRefGoogle Scholar
  38. Discamps, E., & Royer, A. (2017). Reconstructing palaeoenvironmental conditions faced by Mousterian hunters during MIS 5 to 3 in southwestern France: A multi-scale approach using data from large and small mammal communities. Quaternary International, 433, 64–87.CrossRefGoogle Scholar
  39. Discamps, E., Jaubert, J., & Bachellerie, F. (2011). Human choices and environmental constraints: deciphering the variability of large game procurement from Mousterian to Aurignacian times (MIS 5-3) in southwestern France. Quaternary Science Reviews, 30(19), 2755–2775.CrossRefGoogle Scholar
  40. d’Errico, F. & Soressi, M. (2018). Manganese-rich fragments from Pech de l’Azé IV: a preliminary analysis. In H. Dibble, S. P. McPherron, P. Goldberg, & D. Sandgathe (Eds.), The Middle Paleolithic Site of Pech de l’Azé IV. Springer.Google Scholar
  41. Farukh, M. A., & Hayasaka, H. (2012). Active forest fire occurrences in severe lightning years in Alaska. Journal of Natural Disaster Science, 33(2), 71–84.CrossRefGoogle Scholar
  42. Genty, D. (2008). Palaeoclimate research in Villars Cave (Dordogne, SW-France). International Journal of Speleology, 37(3), 3.CrossRefGoogle Scholar
  43. Gill, J. L., Williams, J. W., Jackson, S. T., Lininger, K. B., & Robinson, G. S. (2009). Pleistocene megafaunal collapse, novel plant communities, and enhanced fire regimes in North America. Science, 326(5956), 1100–1103.CrossRefGoogle Scholar
  44. Goldberg, P., Dibble, H., Berna, F., Sandgathe, D., McPherron, S. J. P., & Turq, A. (2012). New evidence on Neandertal use of fire: examples from Roc de Marsal and Pech de l'Azé IV. Quaternary International, 247(1), 325–340.CrossRefGoogle Scholar
  45. Goldberg, P., McPherron, S. P., Dibble, H., and Sandgathe, D. (2018). Stratigraphy, deposits, and site formation. In Dibble, H. L., S. P. McPherron, P. Goldberg, and D. Sandgathe, The Middle Paleolithic Site of Pech de l’Azé IV. Cave and Karst Systems of the World Series (pp. 21–74). Springer.Google Scholar
  46. Goldberg, P., Weiner, S., Bar-Yosef, O., Xu, Q., & Liu, J. (2001). Site formation processes at Zhoukoudian, China. Journal of Human Evolution, 41(5), 483–530.CrossRefGoogle Scholar
  47. Goni, M. F. S., Landais, A., Fletcher, W. J., Naughton, F., Desprat, S., & Duprat, J. (2008). Contrasting impacts of Dansgaard–Oeschger events over a western European latitudinal transect modulated by orbital parameters. Quaternary Science Reviews, 27(11), 1136–1151.CrossRefGoogle Scholar
  48. Gowlett, J. A., & Wrangham, R. W. (2013). Earliest fire in Africa: towards the convergence of archaeological evidence and the cooking hypothesis. Azania: Archaeological Research in Africa, 48(1), 5–30.CrossRefGoogle Scholar
  49. Guadelli, J.-L., & Laville, H. (1988). L'environnement climatique de la fin du Moustérien à Combe-Grenal et à Camiac. In Paléolithique moyen récent et Paléolithique supérieur ancien en Europe. Ruptures et transitions: examen critique des documents archéologiques, (Vol. 3, pp. 43–48, 41 fig., 41 tab.): Mémoires du Musée de Préhistoire d'Ile-de-France. APRAIF éd., NemoursGoogle Scholar
  50. Guérin, G., Discamps, E., Lahaye, C., Mercier, N., Guibert, P., Turq, A., et al. (2012). Multi-method (TL and OSL), multi-material (quartz and flint) dating of the Mousterian site of Roc de Marsal (Dordogne, France): Correlating Neanderthal occupations with the climatic variability of MIS 5-3. Journal of Archaeological Science, 39(10), 3071–3084.CrossRefGoogle Scholar
  51. Guiot, J., De Beaulieu, J., Cheddadi, R., David, F., Ponel, P., & Reille, M. (1993). The climate in Western Europe during the last glacial/interglacial cycle derived from pollen and insect remains. Palaeogeography, Palaeoclimatology, Palaeoecology, 103(1–2), 73–93.CrossRefGoogle Scholar
  52. Henry, A. G. (2017). Neanderthal cooking and the costs of fire. Current Anthropology, 58(S16), S329–S336.CrossRefGoogle Scholar
  53. Heyes, P. J., Anastasakis, K., De Jong, W., Van Hoesel, A., Roebroeks, W., & Soressi, M. (2016). Selection and use of manganese dioxide by Neanderthals. Scientific Reports, 6, 22159.CrossRefGoogle Scholar
  54. Hlubik, S., Berna, F., Feibel, C., Braun, D., & Harris, J. W. (2017). Researching the nature of fire at 1.5 mya on the site of FxJj20 AB, Koobi Fora, Kenya, using high-resolution spatial analysis and FTIR spectrometry. Current Anthropology, 58(S16), S243–S257.CrossRefGoogle Scholar
  55. Hodgkins, J. (2018). Taphonomical and Zooarchaeological Analysis of Bordes’ Excavated Material from Levels I2 and Y-Z. In H. Dibble, S. P. McPherron, P. Goldberg, & D. Sandgathe (Eds.), Excavations at Pech de l’Azé IV: A Middle Paleolithic Site in Southwest France Google Scholar
  56. Hodgkins, J., Marean, C. W., Turq, A., Sandgathe, D., McPherron, S. P., & Dibble, H. L. (2016). Climate-mediated shifts in Neandertal subsistence behaviors at Pech IV and Roc de Marsal (Dordogne Valley, France). Journal of Human Evolution, 96, 1–18.CrossRefGoogle Scholar
  57. Jankowski, N. R. (2018) An Absolute Chronological Framework for Pech IV. In: H. Dibble, S. P. McPherron, P. Goldberg, & D. Sandgathe (Eds.), The Middle Paleolithic Site of Pech de l’Azé IV. Springer.Google Scholar
  58. Jelinek, A. (2013). Neandertal Lithic Industries at La Quina. Tucson: University of Arizona Press.Google Scholar
  59. Julig, P., Long, D., Schroeder, H., Rink, W., Richter, D., & Schwarcz, H. (1999). Geoarchaeology and new research at Jerf al-‘Ajla cave, Syria. Geoarchaeology, 14(8), 821–848.CrossRefGoogle Scholar
  60. Karkanas, P. (2001). Site formation processes in Theopetra Cave: a record of climatic change during the Late Pleistocene and Early Holocene in Thessaly, Greece. Geoarchaeology, 16(4), 373–399.CrossRefGoogle Scholar
  61. Karkanas, P., Rigaud, J. P., Simek, J. F., Albert, R. M., & Weiner, S. (2002). Ash bones and guano: a study of the minerals and phytoliths in the sediments of Grotte XVI, Dordogne, France. Journal of Archaeological Science, 29(7), 721–732.CrossRefGoogle Scholar
  62. Laville, H. (1975). Climatologie et chronologie du Paléolithique en Périgord: étude sédimentologique de dépôts en grottes et sous abris (Vol. 4): Editions du Laboratoire de paléontologie humaine et de préhistoire, Université de Provence, Centre Saint Charles.Google Scholar
  63. Mallol, C., Hernández, C. M., Cabanes, D., Sistiaga, A., Machado, J., Rodríguez, T., et al. (2013). The black layer of Middle Palaeolithic combustion structures. Interpretation and archaeostratigraphic implications. Journal of Archaeological Science, 40(5), 2515–2537.CrossRefGoogle Scholar
  64. Mallol, C., Marlowe, F. W., Wood, B. M., & Porter, C. C. (2007). Earth, wind, and fire: Ethnoarchaeological signals of Hadza fires. Journal of Archaeological Science, 34(12), 2035–2052.CrossRefGoogle Scholar
  65. Mcpherron, S. P., & Dibble, H. L. (1999). The lithic assemblages of Pech de l'Azé IV (Dordogne, France). Préhistoire Européenne, 15, 9–43.Google Scholar
  66. McPherron, S. P., Dibble, H. L., Goldberg, P., Lenoir, M., Sandgathe, D., & Turq, A. (2012). De Combe Grenal à Pech de l'Azé IV: L'évolution des Méthodes de Fouilles de François Bordes. François Bordes et la Préhistoire. Actes du colloque de Bordeaux en, 2009, 65–77.Google Scholar
  67. McPherron, S. P., Dibble, H. L., Sandgathe, D., Goldberg, P., & with contributions by S. Lin and A. Turq. (2018). The Lithic Assemblages. In H. L. Dibble, S. P. McPherron, P. Goldberg, & D. Sandgathe (Eds.), The Middle Paleolithic Site of Pech de l’Azé IV (pp. 117–219). Berlin: Springer, Cave and Karst Systems of the World Series.CrossRefGoogle Scholar
  68. Miller, C. E. (2015). A tale of two Swabian caves. Geoarchaeological investigations at Hohle Fels and Geißenklösterle. Tübingen: Kerns Verlag.Google Scholar
  69. Montoya, E., Rull, V., Stansell, N. D., Abbott, M. B., Nogué, S., Bird, B. W., et al. (2011). Forest–savanna–morichal dynamics in relation to fire and human occupation in the southern Gran Sabana (SE Venezuela) during the last millennia. Quaternary Research, 76(3), 335–344.CrossRefGoogle Scholar
  70. Mooney, S., Harrison, S., Bartlein, P., Daniau, A.-L., Stevenson, J., Brownlie, K., et al. (2011). Late quaternary fire regimes of Australasia. Quaternary Science Reviews, 30(1), 28–46.CrossRefGoogle Scholar
  71. Morin, E., Delagnes, A., Armand, D., Castel, J.-C., & Hodgkins, J. (2014). Millennial-scale change in Archaeofaunas and their implications for Mousterian lithic variability in Southwest France. Journal of Anthropological Archaeology, 36, 158–180.CrossRefGoogle Scholar
  72. Niven, L. (2013). A diachronic evaluation of Neanderthal cervid exploitation and site use at Pech de l’Azé IV, France. In: Zooarchaeology and Modern Human Origins (pp. 151–161). Springer.Google Scholar
  73. Niven, L. (2018). Zooarchaeological Analysis of the Assemblage from the 2000–2003 Excavations. In: H. Dibble, S. P. McPherron, P. Goldberg, & D. Sandgathe (eds.), Excavations at Pech de l’Azé IV: A Middle Paleolithic Site in Southwest France. Google Scholar
  74. Niven, L., Steele, T. E., Rendu, W., Mallye, J.-B., McPherron, S. P., Soressi, M., Jaubert, J., & Hublin, J.-J. (2012). Neandertal mobility and large-game hunting: The exploitation of reindeer during the Quina Mousterian at chez-Pinaud Jonzac (Charente-maritime, France). Journal of Human Evolution, 63(4), 624–635.CrossRefGoogle Scholar
  75. Pasto, I., Allue, E., & Vallverdu, J. (2000). Mousterian Hearths at Abric Romani, Catalonia (Spain). Neanderthals on the Edge Oxford, Oxbow Books (pp. 59–67).Google Scholar
  76. Pop, E., Kuijper, W., van Hees, E., Smith, G., García-Moreno, A., Kindler, L., et al. (2016). Fires at Neumark-Nord 2, Germany: an analysis of fire proxies from a last interglacial Middle Palaeolithic basin site. Journal of Field Archaeology, 41(5), 603–617. Scholar
  77. Pruetz, J. D., & Herzog, N. M. (2017). Savanna chimpanzees at Fongoli, Senegal, navigate a fire landscape. Current Anthropology, 58(S16), S337–S350.CrossRefGoogle Scholar
  78. Pruetz, J. D., & LaDuke, T. C. (2010). Brief communication: Reaction to fire by savanna chimpanzees (Pan troglodytes verus) at Fongoli, Senegal: Conceptualization of “fire behavior” and the case for a chimpanzee model. American Journal of Physical Anthropology, 141(4), 646–650.Google Scholar
  79. Reille, M., & De Beaulieu, J. (1990). Pollen analysis of a long upper Pleistocene continental sequence in a Velay maar (Massif Central, France). Palaeogeography, Palaeoclimatology, Palaeoecology, 80(1), 35–48.CrossRefGoogle Scholar
  80. Richter, D., Hublin, J.-J., Jaubert, J., McPherron, S. P., Soressi, M., & Texier, J.-P. (2013). Thermoluminescence dates for the Middle Palaeolithic site of Chez-Pinaud Jonzac (France). Journal of Archaeological Science, 40(2), 1176–1185. Scholar
  81. Rodríguez-Cintas, Á., & Cabanes, D. (2017). Phytolith and FTIR studies applied to combustion structures: the case of the middle Paleolithic site of El Salt (Alcoy, Alicante). Quaternary International, 431, 16–26.CrossRefGoogle Scholar
  82. Roebroeks, W., & Villa, P. (2011). On the earliest evidence for habitual use of fire in Europe. Proceedings of the National Academy of Sciences of the United States of America, 108(13), 5209–5214.CrossRefGoogle Scholar
  83. Romps, D., Seeley, J., Vollaro, D., & Molinari, J. (2014). Projected increase in lightning strikes in the United States due to global warming. Science, 246, 851–854.CrossRefGoogle Scholar
  84. Rowlett, R. M. (2000). Fire control by Homo erectus in East Africa and Asia. Acta Anthropologica Sinica, 19(5uppl), 198–208.Google Scholar
  85. Sandgathe, D., Goldberg, P., Dibble, H., McPherron, S. P., Turq, A., & Schowortz, S. (2008). Roc de Marsal (Campagne-de-Bugue, Dordogne). Rapport d'Operation pour l'Annee 2008. Report on file with the Service Regional de l'Archaeologie, Bordeaux.Google Scholar
  86. Sandgathe, D., & Berna, F. (2017). Symposium introduction to fire and the genus Homo. Current Anthropology, 58(16), S165–S174.CrossRefGoogle Scholar
  87. Sandgathe, D. M. (2017). Identifying and describing pattern and process in the evolution of Hominin use of fire. Current Anthropology, 58(16), S360–S370.CrossRefGoogle Scholar
  88. Sandgathe, D. M., Dibble, H., Goldberg, P., McPherron, S. P., & Turq, A. (2004). Roc de Marsal (Campagne-de-Bugue, Dordogne). Rapport d’Opération pour l’Année 2004. Report on file with the Service Régional de l’Archéologie, Bordeaux.Google Scholar
  89. Sandgathe, D. M., Dibble, H. L., Goldberg, P., McPherron, S. P., Turq, A., Niven, L., et al. (2011). On the Role of Fire in Neandertal Adaptations in Western Europe: Evidence from Pech de l'Azé and Roc de Marsal, France. PaleoAnthropology, 216–242Google Scholar
  90. Schmidt, O. (2006). Wood and tree fungi: biology, damage, protection, and use. Berlin: Springer Science & Business Media.Google Scholar
  91. Shimelmitz, R., Kuhn, S. L., Jelinek, A. J., Ronen, A., Clark, A. E., & Weinstein-Evron, M. (2014). ‘Fire at will’: the emergence of habitual fire use 350,000 years ago. Journal of Human Evolution, 77, 196–203.CrossRefGoogle Scholar
  92. Sorensen, A., Roebroeks, W., & van Gijn, A. (2014). Fire production in the deep past? The expedient strike-a-light model. Journal of Archaeological Science, 42, 476–486. Scholar
  93. Sorensen, A. (2017a). On the relationship between climate and Neandertal fire use during the last glacial in south-west France. Quaternary International, 436, 114–128.CrossRefGoogle Scholar
  94. Sorensen, A. (2017b). MTA bifaces used as percussive fire-making tools by late Neandertals. Presented to annual meeting of the European Society for the Study of Human Evolution, Leiden.Google Scholar
  95. Soressi, M., & d'Errico, F. (2007) Pigments, gravures, parures: les comportements symboliques controversés des Néandertaliens. Éditions du CTHS.Google Scholar
  96. Soressi, M., Armand, D., D'errico, F., Jones, H., Pubert, E., Rink, W. J., Texier, J.-P., & Vivent, D. (2002). Pech-de-l'Azé I (Carsac, Dordogne): Nouveaux travaux sur le Moustérien de tradition acheuléenne. Bulletin De La Société Préhistorique Française, 99, 5–11.CrossRefGoogle Scholar
  97. Soressi, M., McPherron, S. P., Lenoire, M., Dogandzic, T., Goldberg, P., Jacobs, Z., Maigrot, Y., Martisius, N. L., Miller, C. E., Rendu, W., Richards, M., Skinner, M. M., Steele, T. E., Talamo, S., & Texier, J.-P. (2013). Neandertals made the first specialized bone tools in Europe. Proceedings of the National Academy of Sciences, 110(35), 14186–14190.CrossRefGoogle Scholar
  98. Speth, J. D. (2017). Putrid meat and fish in the Eurasian Middle and Upper Paleolithic: are we missing a key part of Neanderthal and Modern Human Diet? PaleoAnthropology, 44–72.Google Scholar
  99. Stahlschmidt, M. C., Miller, C. E., Ligouis, B., Hambach, U., Goldberg, P., Berna, F., et al. (2015). On the evidence for human use and control of fire at Schöningen. Journal of Human Evolution, 89, 181–201.CrossRefGoogle Scholar
  100. Stiner, M. C., Kuhn, S. L., Weiner, S., & Bar-Yosef, O. (1995). Differential burning, recyrstallization, and fragmentation of archaeological bone. Journal of Archaeological Science, 22, 223–237.CrossRefGoogle Scholar
  101. Tzedakis, P. (1994). Vegetation change through glacial-interglacial cycles: a long pollen sequence perspective. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 345(1314), 403–432.CrossRefGoogle Scholar
  102. Vallverdú, J., Alonso, S., Bargalló, A., Bartrolí, R., Campeny, G., Carrancho, Á., et al. (2012). Combustion structures of archaeological level O and mousterian activity areas with use of fire at the Abric Romaní rockshelter (NE Iberian Peninsula). Quaternary International, 247(1), 313–324.CrossRefGoogle Scholar
  103. van der Kaars, S., Wang, X., Kershaw, P., Guichard, F., & Setiabudi, D. A. (2000). A Late Quaternary palaeoecological record from the Banda Sea, Indonesia: patterns of vegetation, climate and biomass burning in Indonesia and northern Australia. Palaeogeography, Palaeoclimatology, Palaeoecology, 155(1), 135–153.CrossRefGoogle Scholar
  104. Verardo, D. J., & Ruddiman, W. F. (1996). Late Pleistocene charcoal in tropical Atlantic deep-sea sediments: climatic and geochemical significance. Geology, 24(9), 855–857.CrossRefGoogle Scholar
  105. Veraverbeke, S., Rogers, B., Goulden, M., Jandt, R., Miller, C., Wiggins, E., & Randerson, J. (2017). Lightning as a major driver of recent large fire years in North Amercian Boreal forests. Nature Climate Change, 7, 529–534.CrossRefGoogle Scholar
  106. Vidal-Matutano, P. (2016). Firewood and hearths: Middle Palaeolithic woody taxa distribution from El Salt, stratigraphic unit Xb (Eastern Iberia). Quaternary InternationalGoogle Scholar
  107. Vogel, J. C., & Waterbolk, H. T. (1967). Groningen radiocarbon dates vii. Radiocarbon, 9(1), 107–155.CrossRefGoogle Scholar
  108. Wein, R. W. (1976). Frequency and characteristics of arctic tundra fires. Arctic, 1976, 213–222.Google Scholar
  109. Weiner, S., Xu, Q., Goldberg, P., Liu, J., & Bar-Yosef, O. (1998). Evidence for the use of fire at Zhoukoudian, China. Science, 281(5374), 251–253.CrossRefGoogle Scholar
  110. Woillard, G. M. (1978). Grande pile peat bog: a continuous pollen record for the last 140,000 years. Quaternary Research, 9(1), 1–21.CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  • Harold L. Dibble
    • 1
    • 2
    • 3
    Email author
  • Dennis Sandgathe
    • 4
    • 5
  • Paul Goldberg
    • 6
    • 7
    • 8
  • Shannon McPherron
    • 2
  • Vera Aldeias
    • 2
    • 9
  1. 1.Department of AnthropologyUniversity of PennsylvaniaPhiladelphiaUSA
  2. 2.Department of Human EvolutionMax Planck Institute for Evolutionary AnthropologyLeipzigGermany
  3. 3.Institute for Human Origins, School of Human Evolution and Social ChangeArizona State UniversityTempeUSA
  4. 4.Department of Archaeology and Human Evolution Studies ProgramSimon Fraser UniversityBurnabyCanada
  5. 5.Museum of Archaeology and AnthropologyUniversity of PennsylvaniaPhiladelphiaUSA
  6. 6.Centre for Archaeological Science, School of Earth & Environmental SciencesUniversity of WollongongWollongongAustralia
  7. 7.Institute for Archaeological SciencesEberhard Karls Universitat TübingenTübingenGermany
  8. 8.Department of ArchaeologyBoston UniversityBostonUSA
  9. 9.Interdisciplinary Center for Archaeology and Evolution of Human Behavior (ICArEHB)Universidade do AlgarveFaroPortugal

Personalised recommendations