Ethnographic data, prehistoric hunter-gatherers and archaic hominins
The use of ethnographic analogy (here defined very broadly as ‘the comparison of different things, settings and practices that share certain properties so as to infer or imply other non-observable commonalities’; Lane 2014:105) has a long history in archaeology. The use of analogy is especially prominent in prehistory, where the comparative richness of the ethnographic database of recent small-scale societies contrasts with an often-underwhelming archaeological record. After decades of discussion on the relative strengths and weaknesses of different types of analogy, and the validity of analogies in archaeological reasoning (eg Gould & Watson 1982; Stahl 1993; Wylie 1985), current consensus views ethnographic analogy as neither universally appropriate nor inappropriate, but something whose validity should be assessed on a context-by-context basis (Currie 2016). In any event, the use of ethnographic analogy in archaeological interpretation is largely inevitable, particularly in early prehistory (Palaeolithic and Mesolithic).
Nonetheless, calls for caution in the use of ethnographic analogy should be taken seriously. For prehistoric hunter-gatherers these typically take two forms. First is the relevance of recent foragers as a source of analogies. Ethnographic hunter-gatherers are largely restricted geographically to marginally productive areas (cf Porter & Marlowe 2007) and are enmeshed in modern world systems (Headland & Reid 1989; Kent 1992; Layton 2001). The question then is how typical they are of the hunter-gatherer adaptation more broadly (if such a thing exists) and whether their lack of cultural ‘purity’ prevents them from being relevant for understanding prehistoric foragers. When the ‘modern’ adaptation emerged (ie that similar to those of ethnographically documented groups), and how this would be recognised archaeologically, has been a particularly important line of enquiry, with many arguing that it only emerged with Homo sapiens in the late Pleistocene (Kuhn & Stiner 2001; Kusimba 2005; Pettitt 2014) or early Holocene (Foley 1988; Richerson & Boyd 2013:292–302).
Secondly, there is the issue of implementation. A key concern is to avoid replicating the ethnographic present in the past; the action of ‘affirming the consequent’ (Gould 1980:29). One possible outcome of this is the inability to recognise unique cases (or those which have no recent counterpart) in the archaeological record, thereby reducing our understanding of the diversity of past societies and practices. The challenge is how to draw on these data in ways which allow for, rather than obscure, any differences between archaeological and ethnographic hunter-gatherers (Wobst 1978; see also Finlayson & Warren 2017 and papers therein).
The relevance of recent foragers to prehistoric hunter-gatherers, and the challenge of imagining ways of being a hunter-gatherer which differ from those seen ethnographically, are compounded when applied to archaic hominins (non-Homo sapiens). Here, we have the additional challenge of studying hominins who are biologically distinct enough from recent foragers to belong to a different species. While it is an anthropological truism that humanity has lived by hunting and gathering for 99% of its existence, definitions of ‘hunter-gatherer’ usually include behaviours beyond these economic activities, including a common mode of social organisation (the ‘band’ eg Lee 1992), similarities in world view (eg the ‘foraging mode of thought’; Barnard 2002 or ‘hunter-gatherer sociality’; Ingold 1987), and gendered division of labour (Barnard 2014:52). The archaic hominins of the Lower and Middle Palaeolithic were undoubtedly pre-agriculture, but it is unclear if and how differences in biology and cognition affected both the way in which they hunted and gathered, and their behaviour and social structures, including those which characterise foragers in the ethnographic present.
With these challenges in mind, rather than offering another ‘ethnographic cautionary tale’ this paper tasks itself with answering the question: how do archaeologists use ethnographic data in their research on archaic hominins? How these data are actually used is a starting point for a broader discussion of the potentials and challenges of their use (ie how they should be used). My focus is on the role of ethnographic analogies in recent research on Neanderthals (Homo neanderthalensis) and the identification of key trends in researchers’ use of data from recent hunter-gatherers to interpret the Neanderthal archaeological record.
The Neanderthals are our closest evolutionary relatives who occupied Eurasia from approximately 300,000 years ago until 40,000 years ago, during the Middle Palaeolithic period of the Middle to Late Pleistocene. Since 2010, studies of the Neanderthal genome have confirmed that some Neanderthals and early Homo sapiens interbred (eg Green et al 2010; Kuhlwilm et al 2016) and that Neanderthals therefore played a role in the origins of Homo sapiens. These findings spearheaded the latest in a long line of fluctuating scholastic opinions on Neanderthals which emphasise the similarities between Neanderthals and Homo sapiens and ushered in the current ‘genomic era’ of Neanderthal research (the focus of this study). As the most recent and well-documented archaic hominin, Neanderthals are an excellent case study for discussing both the possible diversity of prehistoric hunter-gatherer adaptations and the implications of biological and cognitive differences between archaic hominins and Homo sapiens for the use of analogies with recent foragers. This paper builds on the recent work of Spikins and colleagues (2017) in considering critically both the current role of ethnographic data in Neanderthal research, the wider implications of this role for interpretations of Neanderthal behaviour, and in providing reflections on best practice.
Characterising the use of ethnographic data in recent Neanderthal archaeological research
How do archaeologists use ethnographic data in their research on Neanderthals? To answer this question a quantitative and qualitative analysis of academic articles published in five international journals (Journal of Anthropological Archaeology [JAA], Journal of Archaeological Science [JAS], Journal of Human Evolution [JHE], Current Anthropology [CA], and Quaternary International [QI]) from the years 2010 to 2017 was carried out. These journals range from primarily archaeological (JAA, JAS) to interdisciplinary publications combining archaeology, palaeoanthropology and quaternary science (CA, QI, JHE), and from quantitative/scientifically (QI, JHE, JAS) to more interpretative/anthropologically oriented (JAA, CA). They form a representative, if not comprehensive, sample of recent Neanderthal research.
Online searches of these journals were made using the key words ‘Neandert(h) al’ and ‘Middle Pal(a)eolithic’. Articles containing these phrases were screened to identify all full-length papers with a primarily archaeological focus on Neanderthals. Articles with a narrow palaeoanthropological or genetic emphasis and/or principally reporting new chronometric/excavation data were excluded, as were articles available online but listed as ‘in press’ in January 2018. The resultant database contained 241 articles (Appendix: supplementary data). Each of these 241 articles was read to collect data on five key questions: 1) Do the authors draw on ethnographic data? 2) If ethnographic data are used, what aspects of Neanderthal behaviour are the focus? (each article was assigned up to three of the classifications listed in
Summary of the results of the analysis of the frequency and use of ethnographic data in published articles on Neanderthal archaeology in five key international journals between the years 2010–2017. The results of questions 2–5 are limited to those articles which draw upon ethnographic data in their analysis or discussion (n=88). The categories presented are not always mutually exclusive. For complete data, see Appendix: supplementary data
|Number of articles|
|Q1: Do the authors draw on ethnographic data?|
|Q2: What aspects of Neanderthal behaviour are the focus? (up to three classifications per article)|
|Taphonomy and spatial organisation||23|
|Q3: How are ethnographic data used?|
|Illustrative (ie neat ethnographic examples to ‘fill in the gaps’, often for narrative purposes)||27|
|General comparative – substantiation (ie used to substantiate a claim or justify a method)||48|
|General comparative – hypothesis testing/search for differences (ie used as starting point for explicit testing or examination of archaeological data)||23|
|Q4a: How specific is the authors’ reference to ethnographic data?|
|Unspecified reference to ‘ethnographic record’ or ‘ethnographic data’||67|
|Use of Binford’s (2001) Frames of Reference dataset||5|
|Reference to specific ethnographic groups||28|
|Q4b: How is the relevance of their choice and use of ethnographic data justified?|
|Shared classification of ‘hunter-gatherers’ (either implicit or explicitly stated)||65|
|Environment (similarities in prevailing climatic/environmental conditions)||14|
|Subsistence (similarities in diet)||2|
|Availability of data||2|
|Q5: Are the difficulties and limitations of the use of ethnographic data acknowledged?|
While every attempt has been made to provide a fair and objective assessment of the use of ethnographic data in these articles, there are some obvious caveats. To maximise the database, the entries are not controlled for authorship/research group, and the varying number of published outputs of individuals or groups with specific approaches and research interests will affect overall trends. Similarly, the journals sampled publish exclusively in English (although researchers based at institutions in non-English speaking countries are well represented). In terms of chronology, it is necessary to note that the draft of the Neanderthal genome was published in May 2010. While the database includes articles published throughout 2010, those published before May do not belong to the genomic era in the strictest sense. Given publication lags, it is likely that some papers published >1 year after this date did not consider these genetic findings in the interpretation and discussion of their data (although such papers would be difficult to identify with certainty). Despite these limitations, the analysis presented here is for the most part systematic and suitable for the examination of the main trends in the use of ethnographic data in recent Neanderthal archaeological research.
Key trends in the use of ethnographic data in genomic era Neanderthal research
Of the 241 articles identified through the literature search, slightly over one third drew on ethnographic data in some way in their analysis (n=88, 36%). The rest of the results refer primarily to these eighty-eight articles and are summarised in
The behavioural focus of each article was identified according to the categories listed in
Ethnographic data are most commonly used for comparative purposes to substantiate or justify a claim (along the lines of ‘often hunter-gatherers do/ethnographic data show that […] so Neanderthals should’) or for narrative purposes, taking the form of illustrative examples to ‘fill in the gaps’ of the archaeological record. The use of ethnographic data as a starting point for testing ideas based on the archaeological data cited is rare. However, in many cases, it was difficult to confidently characterise the use of ethnographic data according to the categories listed in
In most cases, engagement with ethnographic data is restricted to unspecified references to ‘the ethnographic record’ or ‘ethnographic data’. Where data on specific groups or populations are cited, researchers draw on a broad range of examples from across four continents including Inuit groups (eg Romagnoli 2015), Aboriginal Australians (eg Vaquero et al 2015), the Hadza (Tanzania) (eg Henry 2017), the Ju/’hoan !Kung (Botswana) (eg Rosell et al 2012), and the Ache (Paraguay) (eg Cochard et al 2012)
It is rare for researchers to provide an explicit justification for the relevance to their Neanderthal data of any ethnographic data drawn upon beyond the shared classification of both populations as ‘hunter-gatherers’. Where some justification is provided, the most common reason is the occupation of similar environments. Other reasons included comparable diets (Germonpré et al 2014), and the simple availability of data on specific ethnographic groups (Bocquet-Appel & Degioanni 2013)
It is rare for researchers to acknowledge the potential problems or limitations of the use of ethnographic data to interpret Neanderthal archaeological data.
Breakdown of the Neanderthal behaviours studied by Palaeolithic archaeologists in articles which draw upon ethnographic data in their analysis and discussion. The behavioural categories are not mutually exclusive, and each article (n=88) was assigned up to three categories (meaning that it may appear more than once in the overall counts). the breakdown of Neanderthal behaviours is shown here as both the total number of articles in each behavioural category which drew upon ethnographic data, and as a percentage of the total number of articles in the database concerned with that behaviour. Data from articles published in Journal of Archaeological Science, Journal of Human Evolution, Journal of Anthropological Archaeology, Current Anthropology and Quaternary International between 2010 and 2017. For complete data, see Appendix: supplementary data
The use of ethnographic data is not ubiquitous in Neanderthal archaeological research, and, if the sample analysed here is representative, the majority of articles published in the genomic era do not draw explicitly upon data from recent hunter-gatherers in their analysis. Nonetheless, the use of ethnographic data to interpret the Neanderthal archaeological record has a long and influential history (eg Binford 1973), and the results of this study show that analogies (as defined in the broadest sense) with recent hunter-gatherers are employed by a sizable number of Neanderthal researchers (cf Spikins et al 2017:132).
Overall, this use of ethnographic data is uncritical, and contra to prevailing recommendations which advocate their use as a source of hypotheses for testing ideas and models with archaeological data, rather than to simply ‘fill in the gaps’ or compensate for an impoverished archaeological record (eg Binford 2001; Kelly 2013). There are several reasons why it might be rare for Neanderthal researchers to use ethnographic data in this way. The frequent lack of the relevant archaeological data and resolution in the Middle Palaeolithic with which to test hypotheses are likely partly responsible. However, there are two other factors which I suggest are at play. The first of these is the lack of engagement by many Neanderthal researchers with the wider field of hunter-gatherer studies. This lack of engagement is not unique to Middle Palaeolithic archaeologists, but is likely compounded by the spatial and temporal distance between Neanderthals and ethnographically documented populations: Neanderthals lived too long ago for any type of direct historical analogy to be applicable and there are no recent forager societies in their Eurasian range. In terms of the data presented here, this lack of engagement is best reflected in the overwhelming lack of critical discussion on the limitations of, or potential problems with, the ethnographic data cited, and the limited consideration given to why the data were relevant to the Neanderthal case (‘source-side’ issues: Stahl 1993:235). More broadly, compared to both the later Palaeolithic and the Mesolithic (eg Porr 2001; Jordan 2006; Warren 2017), there is little literature dedicated to critical reflection on the use of analogy and ethnographically derived models in Middle Palaeolithic archaeology.
The second reason is specific to the study of Neanderthals in the genomic era. One outcome of the finding of interbreeding between Neanderthals and early Homo sapiens – bolstered by new archaeological evidence for symbolic and artistic behaviours amongst Neanderthals (Hoffman et al 2018; Peresani et al 2013; Romandini et al 2014), and for new technologies (Hardy et al 2013; Kozowyk et al 2017) – is a current research trend which emphasises the similarities between the two groups (see also Spikins et al 2017). At its extreme, this view claims for the archaeological ‘indistinguishability’ (Wynn et al 2016) of Neanderthals and early Homo sapiens (eg Hayden 2012; Villa & Roebroeks 2014; Zilhão 2014; see Wynn et al 2016 for a refutation). Assuming such strong behavioural similarities arguably reduces the need to articulate the relevance of recent foragers to Neanderthals, and takes for granted the appropriateness of analogues with recent foragers (although as noted earlier, several scholars place the emergence of the ‘modern forager adaptation’ at the end of the Pleistocene/Early Holocene, also excluding the early Upper Palaeolithic Homo sapiens with whom Neanderthals are most frequently compared). However, an investigation of directionality is beyond the scope of this study, and it is therefore difficult to state both how the use of ethnographic data which characterises the current decade differs from past research, and the extent to which recent findings are driving this trend.
Regardless of why, the fact remains that the current use of ethnographic data in Neanderthal archaeology frequently falls into the traps discussed in the introduction which largely replicate the ethnographic record in the past, and do not allow for the possibility of differences between the archaeological and ethnographic cases. These results support those of Spikins et al (2017), who argued for a similar scenario based on the close reading of several research papers, most notably Hayden (2012). This practice is particularly problematic for Neanderthals as it brushes aside some notable differences in biology (and, to a lesser extent, presumably cognition) between Neanderthals and Homo sapiens. These differences likely had an important impact on the social and cultural behaviours we seek to understand. Before considering the impact of biological and cognitive differences between Neanderthals and Homo sapiens on both our interpretation of the Middle Palaeolithic archaeological record and inferences drawn from recent foraging populations, it is necessary to review briefly the evidence for these differences.
Biological and cognitive differences between Neanderthals and Homo sapiens
Neanderthals evolved in Europe during the Middle Pleistocene following the divergence of their lineage with that which led to Homo sapiens. Genetic estimates place this divergence between ~765 and 550 ka (Prüfer et al 2014), with the earliest fossils which show Neanderthal-like features dated to ~430 ka (Arsuaga et al 2014). The identity of the last common ancestor of Neanderthals and Homo sapiens is unclear. Homo heidelbergensis is typically considered the most likely candidate (Stringer 2012), although the young age of several presumed Homo heidelbergensis fossils weaken this hypothesis (Harvati 2016; Manzi et al 2010). The species status of Neanderthals is also debated; whether they are a separate species (Homo neanderthalensis), or a sub-species of Homo sapiens (Homo sapiens neanderthalensis). Here, I follow the lead of recent reviews by Churchill (2014) and White et al (2014) to consider Neanderthals a separate species. Regardless of the position taken on the Neanderthal species question, the fact remains that Neanderthals are both closely related to Homo sapiens, and experienced millennia of a distinct evolutionary trajectory.
Morphologically, the constellation of Neanderthal craniofacial features best distinguishes them from other closely related hominins (Churchill 2014:14). Estimates of Neanderthal endocranial volume suggest brain size (relative to overall body mass) comparable to Homo sapiens as well as larger absolute brain sizes (Robson & Wood 2008). Key cranial differences between Neanderthals and Homo sapiens include the shape of the cranial vault (with Neanderthals characterised by a long and broad shape, compared to the upright and rounded cranial vault of Homo sapiens; Bruner et al 2003), and the robust Neanderthal supraorbital and orbital tori (brow ridges), pronounced mid-face prognathism and corresponding lack of chin. Post-cranially, Neanderthals were on average shorter and stockier than Homo sapiens, with features including a wide trunk, barrel-shaped chest, large muscle attachment sites, and short distal extremities (forearms and legs) (see summary in Harvati 2007); a morphology variously attributed to adaptation to cold glacial conditions (Holliday 1997; Steegman et al 2002) or extremely high activity levels (Franciscus & Churchill 2002; Weaver 2009).
One much contested potential area of biological difference between Neanderthals and Homo sapiens is their respective patterns of growth and development (life history), particularly whether Neanderthals experienced a long childhood phase comparable to that of Homo sapiens. The results of a comparative review suggest that the life history of Neanderthals was consistent with that of Homo sapiens, based on data on body size, brain size, and dental development (Robson & Wood 2008); a finding supported by a recent analysis of the growth pattern of the El Sidrón Neanderthal child (Rosas et al 2017). However, other studies suggest a Neanderthal growth pattern both faster (Hublin et al 2015; Ramírez Rossi & Bermúdez de Castro 2004; Smith et al 2007) and potentially slower (Ponce de León et al 2008) than those of Homo sapiens. Some of these contradictions can likely be explained by using different measures of development (eg dental vs cranial evidence) (Guatelli-Steinberg 2009).
Another area of potential difference is energetic requirements. The large and muscular body mass of Neanderthals has led many researchers to suggest that they had a higher Basal Metabolic Rate (BMR) (the minimum amount of energy required to sustain basic biological functions) than Homo sapiens. This increased BMR – in combination with the increased energetic requirements of combating cold stress, the consumption of high protein diets and inferred high activity levels and higher energetic travel costs (Weaver & Steudel-Numbers 2005) – results in estimates of adult Neanderthal daily total energy expenditure (TEE) up to 10% higher than those for both Upper Palaeolithic and present-day Homo sapiens (Churchill 2014; MacDonald et al 2009; Snodgrass & Leonard 2009; Sorensen & Leonard 2001; cf Heyes & MacDonald 2015).
Data on possible cognitive differences between Homo sapiens and extinct hominins are more elusive. Inferences of Neanderthal cognition based on factors such as brain size and shape are difficult to make, but, when combined with archaeological information on behaviour, provide the most direct data (see Wynn & Coolidge 2012 for a review). Wynn and colleagues (2016) identify several domains of likely cognitive difference between Neanderthals and Homo sapiens as implied by their cranial morphologies, including those relating to motor learning, creative thinking and memory. These differences were likely subtle and are difficult to quantify. Furthermore, as Wynn et al (2016), stress, these differences do not imply value judgement as to which cognitive characteristics are ‘better’ or ‘worse’. Given both the difficulty of assessing cognitive differences, and the historical use of assumptions of inferior Neanderthal cognition (compared to Homo sapiens) as a ‘fall back’ position in archaeology to explain perceived limitations to their behaviour, the rest of this discussion will focus on other aspects of Neanderthal biology. This is not to deny, however, the importance of cognition for our understanding of hominin behaviour, adaptation and evolution. Here I consider briefly the implications of these biological differences between Neanderthals and Homo sapiens with reference to two interrelated domains: demography and mobility.
The relationship between Neanderthal biology and behaviour
The study of Neanderthal demography is notoriously difficult. Nonetheless, most researchers agree that Neanderthals lived in small groups and at low population densities compared to both Upper Palaeolithic Homo sapiens and recent hunter-gatherers (eg Bocquet-Appel & Degioanni 2013; Holliday et al 2014; Mellars & French 2011). One possible explanation for this pattern is taphonomic loss through time. Many archaeological estimates of population sizes and densities rely on the quantification of proxies such as numbers of sites and quantities of material culture, which a priori are less likely to survive the further back in time one goes. Nonetheless, this pattern is consistent enough across multiple lines of evidence (archaeological, genetic, osteological) to suggest that it is at least partly real, and that Neanderthals were indeed characterised by small living groups, overall small population sizes and low population densities.
What role did the Neanderthal biology play in this? Although culture and behaviour are also important, biology is fundamental to all demographic processes. Researchers have long hypothesised as to the impact of the robust Neanderthal physiology and their high energetic demands on their group size, population density, and in particular, their fertility (Snodgrass & Leonard 2009). At the heart of these arguments is the idea that reproduction is energetically expensive and requires additional calories above normal metabolic requirements both to support the pregnancy and the subsequent lactation required to feed the child (Butte & King 2005). The overall energy balance of a woman also plays an important role in her ability both to become pregnant, and to complete a successful pregnancy (Ellison 2001). In combination with already high energetic requirements, these additional constraints would likely have resulted in low Neanderthal fecundity (ability to conceive) and low fertility levels (Churchill 2014). The evidence also suggests high Neanderthal mortality rates (Trinkaus 1995) and possibly also low life expectancy (Caspari & Lee 2004). This combination of low fertility and high mortality – driven by physiological constraints and life history traits – means that Neanderthals had difficulty maintaining viable populations, let alone experiencing population growth. The result would have been overall small population sizes, low population densities and high susceptibility to local extinctions (see calculations by Caspari et al 2017; Nakahashi et al 2018; Sørensen 2011).
Neanderthal mobility and land use
How might Neanderthal biology have influenced their land use and mobility patterns? Verpoorte and colleagues’ (Verpoorte 2006; MacDonald et al 2009; Roebroeks & Verpoorte 2009) model of the effect of the robust Neanderthal morphology on these elements of behaviour provides one example. Assuming foraging from a central place across a homogenous environment, the higher energetic requirements of Neanderthals would have reduced their effective foraging radius (the distance from camp to the location of the food resource at which the required amount of energy is equal to the net return of resources at that distance). All other things being equal, the same site would be a feasible place of residence for Neanderthals for a shorter time than Homo sapiens, meaning more residential moves (but of a shorter distance) per year for Neanderthals. A comparative study of the land-use strategies of Neanderthals and early Homo sapiens in the Levant provides broad support for this model, with Neanderthal site exploitation territories (foraging ranges) up to 20% smaller than those of Homo sapiens (Henry et al 2017). Neanderthal high energetic costs might also have influenced overall mobility strategies. Churchill (2014:Chapter 10; Churchill et al 2016) suggests that these costs, in combination with other aspects of Neanderthal dietary ecology (heavy reliance on meat) and technological repertoire (lack of transportation technology), would have favoured a primarily residential (moving people to resources) rather than a logistical (moving resources to people) mobility and settlement pattern.
More broadly, aspects of biology may explain divergences between the Neanderthal archaeological record and expectations based on cross-cultural patterns seen in recent hunter-gatherers. Ethnographic foragers display clear cross-cultural patterning across a range of behaviours driven by adaptations to environmental conditions. Mid-to-high latitude hunter-gatherers, who most closely approximate Neanderthals in terms of environments inhabited, are characterised by an emphasis on hunted rather than gathered resources, large home ranges and predominantly logistical mobility strategies, a reliance on storage, and diverse and complex toolkits (see, for example; Binford 2001; Roscoe 2003; Torrence 1989). Neanderthals do not adhere to these cross-cultural expectations (Bocquet-Appel & Tuffreau 2009; Hardy & Moncel 2011; Hoffecker 2002:131–137; Kuhn 2011). One possible explanation for this is the lower population density of Neanderthals which caused them to adapt differently to prevailing environmental conditions. Churchill (2014:53–54) postulates that the lack of complex Neanderthal subsistence technology (an adaptation designed to mitigate the increased levels of risk in resource acquisition at higher latitudes, both in terms of reduced encounters and severity of resource failure) means that, unlike ethnographic foragers, they responded to the challenges of obtaining food in these ecological conditions in other, non-technological ways (eg altering the size and composition of their foraging groups, and/or foraging ranges). However, both the plausibility of these adaptations and the extent to which biological differences are an adequate explanation is unclear; Middle Palaeolithic Homo sapiens also do not correspond to these latitudinally-driven patterns (Kuhn & Stiner 2001), which may be exclusive to Holocene environments (Osborn 1999).
Implications for ethnographic analogy
The interactions between Neanderthal biology and behaviour discussed above have clear implications for the use of ethnographic analogy in Neanderthal archaeological research. Alongside the important social and behavioural effects of vastly different historical trajectories between recent and prehistoric hunter-gatherers, the possible effects of these interactions need to be recognised and taken into account. This is rarely the case: only four articles included in the database implicitly addressed the complicating factor of biological differences between Neanderthals and recent foragers (Browne & Wilson 2011; Castel et al 2017; Henry et al 2017; Nicholson 2017).
Translating this recognition into direct recommendations for how archaeologists should use ethnographic data differently in Neanderthal contexts is difficult. As discussed above, one recommendation which applies more broadly is using these data as a starting point to test hypotheses about prehistoric foragers; a practice which is often conducted with the paradigm of Human Behavioural Ecology (Humphreys 2007; Kelly 2013). Frustratingly, the archaeological data required to use ethnographic analogy in this way are rarer the further back in time one goes, where both the necessity and benefit of such a strategy is arguably greatest. Nonetheless, several articles included in this database used ethnographic data in this way (eg Dusseldorp 2012; Henry et al 2014; Hockett 2012; Salazar-García et al 2013) indicating that the challenges of archaeological data resolution and availability are not insurmountable in all Middle Palaeolithic contexts.
The overall recommendation presented here is more critical engagement with the ethnographic data used. One immediate concern is the direct transplant of absolute values from the ethnographic record to Neanderthal contexts. Continuing with the example of demography, as the evidence suggests that Neanderthals likely lived in smaller groups than recent hunter-gatherers, the transference of group size estimates from recent foragers to Neanderthals (eg Hayden 2012) should be questioned, at least without additional justification. One response to this has been the use of data from those ethnographic groups living at the lowest observed densities and/or minimum group size estimates when generating expectations for Neanderthal populations (eg Binford 2007). Similarly, the explicit application of criteria of suitability to the ethnographic data selected, and the removal of cases deemed anachronistic or otherwise unsuitable, should also be considered. Of the articles included in the dataset analysed here, only four applied such criteria. The idea that certain features of a recent hunter-gatherer group’s adaptation could be removed to reveal the ‘pure’ hunter-gatherers is difficult to justify, implying (incorrectly) that forager cultures are essentially static in the absence of outside influences (Sassaman & Holly 2011; Layton 2001:292). However, the acknowledgement that some data are more relevant than others is warranted. For example, when addressing Neanderthal demography and/or mobility, recent foragers who are primarily sedentary are likely to be of reduced comparative value (see Holliday et al 2014; Pearce & Moutisou 2014).
Critical use of ethnographic data also involves engaging more with the specifics of how these data were collected. While it is often tempting to defer to ethnographic data, we need to bear in mind the many assumptions, limitations and possible sources of error involved in both their collection and analysis. For example, both Howell (1979), and Hill and Hurtado (1996) discuss at length the challenges of accurate demographic data collection in ethnographic contexts. More broadly, Bird-David (2018) argues that anthropologists have considered inadequately the effects of the small size of hunter-gatherer societies on both other elements of forager cultures and their interpretation of these societies, effects which were likely more pronounced in the smaller societies of prehistory and which archaeologists should take into account. Along these lines, Spikins and colleagues (2017:137) suggest that the implications of Neanderthal biology to their demography and settlement patterns combine to result in an ‘intimate scale of Neanderthal sociality’, in which the local focus of Neanderthal social structures and interactions contrasts with that of both archaic Homo sapiens and recent hunter-gatherers.
Neanderthal demography and mobility are two domains where models and data from ethnography are particularly prevalent. They are also domains where we might expect interactions between biology and behaviour to be particularly important in generating the patterns seen, and in which the biological differences between Neanderthals and Homo sapiens (including recent foragers) mean that the uncritical use of ethnographic analogy may be especially problematic. What about other behaviours, including those (discussed in the introduction) which characterise foragers in the present day? One such example is the gendered division of labour seen among most ethnographic foragers (men hunt, women gather). Whether or not Neanderthals shared the gendered division of labour is an ongoing debate (Kuhn & Stiner 2006). While the direct evidence for or against the presence of this division is ambiguous, small Neanderthal group size may have played a key role. If technological and economic skills are transmitted culturally between individuals, the division of tasks by gender, and the maintenance of this division, requires twice as many people to ensure the successful transmission of cultural knowledge. This division might have been unfeasible if group sizes were too small and/or mortality was high (Nakahashi 2017), factors which, as discussed above, are strongly influenced by Neanderthal biology. In terms of lithic technology, while fracture mechanics of flint adhere to uniformitarian principles, the interplay between Neanderthal cognition, biology and behaviour (including the effect of these on social structures and interactions) undoubtedly affected the important processes of manufacture, use and discard. Ultimately, when considering archaic hominins, behaviour and biology should not be separated (particularly the impact of differing biology and cognition on behaviour: Foley 2002), nor subsumed under broad ethnographic analogies.
I do not mean to imply that there are no benefits to the prevailing use of ethnographic data in Neanderthal archaeological research. The Middle Palaeolithic record is undoubtedly patchy, and analogies are good tools to think with, both in terms of interpreting our evidence, and in reminding us that people created it. Furthermore, the use of ethnographic analogies and frameworks is largely inevitable and firmly embedded in Palaeolithic research traditions. Binford’s (1980) model of residential and logistical hunter-gatherer mobility which was referenced in twenty-five of the papers analysed, is a case in point. Warren (2017:153) argues that this model is often treated as ‘received wisdom’ in hunter-gatherer archaeology, and the irony of my uncritical use of the residential/logistical mobility continuum to describe Neanderthal mobility practices, while simultaneously discussing the shortcomings of such use, is not lost. The embeddedness of such models demonstrates the importance of ongoing critical evaluation of our engagement with ethnographic data for both Neanderthal and wider prehistoric hunter-gatherer research.
This analysis of the use of ethnographic data in recent Neanderthal archaeological research was designed to be expository, not accusatory. Furthermore, the results might appear unsurprising to many archaeologists, and the recommendations here are not new. However, it continues to be the case that ‘one of the greatest challenges in Palaeolithic archaeology is imagining ways of being human that are not like those of recent hunter-gatherers […] it is important to avoid […] imagining early hominin groups as essentially recent foragers lacking a few key elements’ (Kuhn & Clark 2015:15). It is difficult to categorically state whether (and to what extent) the prevailing use of ethnographic data results in a weakening of our understanding of Neanderthals. The use of ethnographic data in ways which serve to replicate the present in the past does, however, serve to weaken our understanding of the diversity of hunter-gatherers. Remaining forager populations are declining, and if we seek to examine the diversity of people who live in small-scale societies and subsist primarily on wild resources, we need to turn to the archaeological record, and the greater geographical coverage and insight into long-term adaptations which it provides. Neanderthals represent an intriguing aspect of this diversity and have the potential to offer new insights into what it means to be a hunter-gatherer.