2. Innovation, Symbolism and Modernity

In order to identify modern behaviour in archaeological traces, palaeolithic archaeology has devised inventories of archaeologically visible behaviours held to be diagnostically modern, the so-called modernity trait list. There is no universally agreed list of diagnostically modern behavioural traits, but since Mellars' first systematic formulation (Mellars 1973), they have been modelled essentially on the archaeology of the European Upper Palaeolithic, and especially the Aurignacian, which appeared in Europe some 45 kya and persisted until around 30 kya. Consequently elements such as a formally standardised and typologically diverse toolkit, blade and bladelet technology, the use of bone, antler and ivory as raw materials, the long-distance transport or exchange of materials, the systematic and effective hunting of dangerous game, and symbolic material culture, have all been regarded as important markers of modernity (e.g. McBrearty and Brooks 2000, 492-3; Klein 2008, table 1).

Today there is wide agreement among archaeologists working in this field that innovativeness and symbolism are the critical and distinctive 'modern' cognitive competencies, and that the behaviours included in the modernity trait list are to be understood as expressions of those competencies. The centrality of the cognitive capacity for symbolism to the modern mind has been argued by Chase and Dibble (1987), Pfeiffer (1982), Byers (1994), Deacon (1997), Deacon and Wurz (2001), Henshilwood and Marean (2003) and Henshilwood and Dubreuil (2011), among others; indeed, Henshilwood and Marean (2003, 635) claim explicitly that symbolism is the sole unique and defining component of modern behaviour. McBrearty and Brooks (2000, 492-3) on the other hand see cognitive modernity as rooted in four critical underlying cognitive abilities: abstract thought, planning depth, innovativeness and symbolism. Mellars (1973; 1989) and Klein (2008, table 1) both identify innovativeness, as illustrated by the relatively rapid turnover of archaeological cultures in the 30,000 year time span of the European Upper Palaeolithic, as a core feature of modernity.

Although coherent repertoires or 'packages' of modern behaviours seem to crystallise only from the beginning of the African Late Stone Age (LSA) some 50 kya, and from the onset of the European Upper Palaeolithic from 45 kya, instances of innovative modern behaviour, including symbolism, occur sporadically in the African Middle Stone Age (MSA) from as long ago as 280 kya (McBrearty and Brooks 2000; Henshilwood and Dubreuil 2011). Examples include personal ornaments and other symbolic objects (e.g. Bouzouggar et al. 2007; Texier et al. 2010); bone harpoons (Yellen et al. 1995); backed, microlithic and geometric lithics (Singer and Wymer 1982; Volman 1984; Barham 2002); and the subsistence exploitation of marine and aquatic resources (Marean et al. 2007). Most striking is Blombos Cave, South Africa, where MSA levels dated to 78 kya have produced numerous pierced marine shells that appear to have been used as 'jewellery' (d'Errico et al. 2005), bone awls (d'Errico and Henshilwood 2007) and fifteen pieces of ochre bearing geometrical incisions, some of which might date to 100 kya (Henshilwood et al. 2009).

The biological-essentialist school is divided on the interpretation of this archaeological evidence along two separate but connected lines. Deacon (1997), McBrearty and Brooks (2000), Deacon and Wurz (2001), Henshilwood and Marean (2003) and Henshilwood and Dubreuil (2011), for example, embrace a broad conception of what constitutes symbolism and regard artefacts, or aspects of artefacts, that cannot be explained in terms of utilitarian function as symbolic; hence their willingness to accept perforated marine shells and geometric engravings as symbolic in character. On the basis of this broad conception of symbolic material culture these authorities place the origin of the modern mind in Homo sapiens of the African Middle Stone Age, who are therefore to be understood as having been cognitively equivalent to living people as early as 77 kya (Henshilwood and Dubreuil 2011) or even as early as 280 kya (McBrearty and Brooks 2000). The inference is that a gradual and uneven cultural accumulation of modern behaviour ensued from an early evolution of the modern human brain, and that by some 50 kya this trajectory had reached a critical adaptive threshold that enabled our species to expand 'Out of Africa' and outcompete archaic hominins to extinction. Others have rejected this claimed cognitive identity between Middle Stone Age and living Homo sapiens, and have insisted that only with the African Late Stone Age and European Upper Palaeolithic after 50 kya does material culture unequivocally indicate modern cognitive capacities for symbolism and syntactical language (e.g. Klein 1995; 1998; 2008; Wynn and Coolidge 2007; Botha 2008; 2009; 2010; Coolidge and Wynn 2009). A critical component of this position is that symbolism is understood more narrowly to include only artefacts that show an arbitrary relationship between signifier and signified; Middle Stone Age perforated shell beads and geometric engravings are dismissed as indexes, rather than as symbols, of group affiliation (e.g. Coolidge and Wynn 2011, 380-1), and thus as having no necessary association with modern cognitive capacities for symbolism and language.

These are certainly important disagreements, but the two positions share critical presumptions in common. Firstly, all protagonists in this particular debate understand symbolism as a conscious and deliberate communication of meanings. Coolidge and Wynn (2011, 380), for example, reject even the tattoos, body piercings and jewellery sported by their students as non-symbolic since the students cannot tell them precisely what those things mean. This rather cybernetic conception of symbolism as flows of information within and between social or cultural entities is far too constrictive, and fails entirely to consider the advances in understanding of material culture as an active participant in social life achieved in archaeology and anthropology over the last thirty years. Here, symbolism is taken as referring to the dynamic incorporation of things, people, places and experiences into the construction and reconstruction of being and personhood through fluid mutual relations in social fields, and is thus to be understood as an aspect of personhood as conceived by Fowler (2004). From this perspective much of the incorporation of things into being and personhood operates at a tacit and habitual, rather than at a conscious and discursive, level, and the things incorporated need not necessarily be deliberately 'made' to be so; consequently the broad understanding of symbolism is to be preferred.

Secondly, both positions are strongly 'brain-bound' in that they share the conviction that modern behaviour, and the modern cognition that enables it, is a property of a modern neurophysiology. There is some disagreement as to the particular character of the essentially modern neurophysiology; Coolidge and Wynn (2001; 2005; 2007) favour a mutation that led to enhanced capacities for working memory and phonological storage, Klein (1998; 2008) suggests a mutation that conferred the cognitive capacity for language, while Henshilwood and Dubreuil (2011) argue for a reorganisation and enhanced connectivity of the temporoparietal junction that enabled the development of symbolism through enhancing capacities for understanding the world from the perspectives of others. Interestingly, both Coolidge and Wynn, and Henshilwood and Dubreuil, understand the emergence of the modern neurophysiology as preceding the emergence of modern human behaviour, since the latter only develops through a process of cultural accumulation enabled by the former. It is clear that, despite the sharp disagreements over what constitutes symbolism and the antiquity of modern cognition, both of these positions remain thoroughly within the biological essentialist consensus.

A major problem with this consensus, however, is the growing evidence for innovation and symbolism in the archaeology of the Neanderthals. It has long been known that prismatic blade lithic technology was practised by Neanderthals in northern Germany and in north-eastern France some ninety millennia prior to the onset of the Upper Palaeolithic (Conard 1990; Ameloot-van der Heijden 1993) and then simply disappeared. Analysis of avian faunal remains from Level XI at Bolomor Cave, Spain, demonstrates systematic Neanderthal subsistence on duck some 160 kya (Blasco and Fernández Peris 2009). Birch bark pitch from Middle Palaeolithic horizons at Königsaue (c. 45-50 kya) and Inden-Altdorf (115-128 kya), Germany, testify both to a remarkable Middle Palaeolithic pyrotechnology and to the use of the pitch as an adhesive in the fabrication of composite hafted tools (Grünberg 2002; Pawlik and Thissen 2011). Many more examples could be given.

There is also growing evidence of Neanderthal symbolism. The Chatelperronian 'transitional' industry of France and northern Spain, dating to the Middle-Upper Palaeolithic boundary between 40 and 44 kya (Mellars and French 2011, 625) is associated with Neanderthal skeletal remains at Saint-Césaire (Lévêque and Vandermeersch 1980) and the Grotte du Renne, Arçy-sur-Cure (Hublin et al. 1996). At the Grotte du Renne the Chatelperronian is characterised by well-made, standardised bone artefacts (some bearing apparently symbolic surface incisions) and animal teeth worked into personal ornaments. Several archaeologists have argued that the Grotte du Renne warrants the Neanderthals being admitted as cognitively and behaviourally modern humans (Zilhão and d'Errico 1999; d'Errico 2003; Zilhão et al. 2008; Caron et al. 2011). This claim has been strongly challenged (Mellars 1996; 2005; Higham et al. 2010).

However, recent research has lent weight to this idea as evidence has emerged for symbolic behaviour in unequivocally Middle Palaeolithic contexts. Perforated marine shells, and shells that bear traces of pigment mixtures, have been recovered from Middle Palaeolithic levels at Cueva de los Aviones and Cueva Antón, southern Spain (Zilhão et al. 2010) and, at 50 kya, are broadly contemporaneous with ochre-grinding receptacles from Grotte de Néron, France (Combier 1989) and Cioarei-Borosteni, Romania (Cârciumaru et al. 2002). Remarkable new evidence has also emerged that Neanderthals sometimes used wing feathers and/or claws extracted from raptors and corvids for display purposes; and that this practice was not restricted to late Neanderthals (Peresani et al. 2011; Finlayson et al. 2012; Morin and Laroulandie 2012).

These developments have serious implications for the biological-essentialist conception of human cognitive modernity. Innovative behaviours, including symbolism, emerged tens or even hundreds of millennia before becoming persistent and widespread components of the 'modern' Upper Palaeolithic package (Hovers and Belfer-Cohen 2006), in both African Homo sapiens and in European Neanderthals, but remained insubstantial and localised before 50 kya. Conversely, the archaeology of late Pleistocene colonisation in many regions outside Europe and Africa offers no evidence at all of symbolic objects, or even of artefact style. This is true at Niah Cave, Borneo, first occupied by Homo sapiens between 35 and 44 kya (Barker et al. 2007), while Bar-Yosef (1998) observes that the full 'Upper Palaeolithic' modernity package only appears in the Levant with the Natufian some 13 kya.


Internet Archaeology is an open access journal. Except where otherwise noted, content from this work may be used under the terms of the Creative Commons Attribution 3.0 (CC BY) Unported licence, which permits unrestricted use, distribution, and reproduction in any medium, provided that attribution to the author(s), the title of the work, the Internet Archaeology journal and the relevant URL/DOI are given.

University of York legal statements

File last updated: Mon June 10 2013