Thinking and living the landscape
Rock art and territories among southern Africa’s nomadic hunter-gatherers
p. 295-313
Résumés
This contribution reflects upon the heuristic potential of rock art for the archaeological study of the sociocultural landscapes of southern Africa’s nomadic hunter-gatherers. Firstly, we explore the work of T. Lessen-Erz and the University of Cologne (Germany) on the spatial structure of painted sites and the social landscapes of Brandberg/Daureb (Namibia) during the Late Holocene. Secondly, we highlight the different lines of questioning exposed through current research on the variability of rock art landscapes on the long-term between the end of the Pleistocene and the Late Holocene within the Matobo (Zimbabwe) and its implications for territorial organization and structure.
Penser et vivre le paysage : art rupestre et territoires chez les nomades chasseurscollecteurs d’Afrique australe
Le paysage est une construction éminemment culturelle : il est non seulement un cadre mais également un acteur de la structuration spatiale et temporelle du monde et de la société. Différents espaces métaphysiques et/ou zones d’exploitation physique le structurent. Des lieux le jalonnent, affectés d’une valeur particulière, et dont la fonction et les modalités de fréquentation (accès, temporalité, durée) sont culturellement et socialement signifiants. Ce réseau de lieux reconnus et définis participe activement à l’identité collective et individuelle. Plus qu’une emprise sur le milieu permettant de bénéficier de ressources, l’identification aux paysages comprend un ensemble de relations et ainsi d’obligations — certaines métaphysiques et rituelles — liant les personnes aux lieux. La mobilité contribue activement à cette construction identitaire dans et à travers les lieux : elle en permet l’expérience, à savoir d’une part la perception, d’autre part l’apprentissage, la connaissance et l’appropriation culturelle, sociale et affective.
Les sites d’art rupestre sont une composante majeure des paysages socio-culturels : ce sont des « marques » (traces intentionnelles) réalisées sur l’environnement (i), sur des supports inamovibles (ii) et pérennes (iii) permettant ainsi de fixer l’emplacement de ces marques dans l’espace et dans la durée. L’art rupestre « inscrit » le paysage. En archéologie, les paysages d’art rupestre constituent donc une voie d’accès privilégiée pour appréhender la structuration territoriale des populations, entendue ici comme la définition et l’organisation des espaces fréquentés c’est-à-dire leurs relations spatio-temporelles et les dynamiques sociales associées (flexibilité des formations sociales). Traitant d’un axe de recherche florissant, cette contribution se veut une réflexion sur le potentiel heuristique des paysages d’art rupestre dans les recherches archéologiques sur les chasseurs-collecteurs d’Afrique australe entre fin Pléistocène et Holocène récent, à travers deux exemples considérant respectivement les échelles synchronique et diachronique.
Les paysages de l’art rupestre comme de multiples facettes des paysages sociaux passés
L’iconographie rupestre répond à des motivations socio-culturelles multiples : principalement spirituelles, de diverses natures, souvent associées à des pratiques rituelles. Elle peut aussi relever de motivations séculières (pédagogique, économique) voire anecdotiques. Elle intervient donc dans des contextes sémantiques multiples et hétérogènes. Dans cette optique, production et utilisation d’images peuvent être fondamentalement considérées comme des agents dans la fabrication, transmission et régulation d’informations entre individus et entre groupes sociaux, notamment dans la négociation des identités sociales (Information Exchange Theory). Leur rôle est donc lié au contexte social (personnes présentes et activités associées), et corrélativement au lieu et au moment. Sous cet angle anthropologique, les sites d’art rupestre peuvent alors être envisagés dans leur statut au sein de la structuration territoriale et des réseaux sociaux. Loin d’être arbitraire, leur implantation résulte d’un dialogue entre « idéologie » — contingences idéelles/spirituelles/symboliques — et « praxis » — nécessités matérielles — qui contribue à faire des paysages d’art rupestre des constructions de l’espace éminemment culturelles et socialement protéiformes.
Les travaux de T. Lenssen-Erz et de l’équipe de l’Université de Cologne sur la structuration spatiale des sites peints du Brandberg/Daureb (Namibie) mettent ainsi en lumière un paysage de l’art rupestre en mosaïque pour les chasseurs-collecteurs qui l’occupèrent à l’Holocène récent (4 000-2 000 BP). Divers dans leur morphologie (rochers, abris) et topographie, les sites sont très largement disséminés avec des densités graphiques très contrastées, d’une poignée de figures à des centaines. Un modèle a été élaboré afin d’interpréter leur fonction socio-culturelle à partir d’un échantillon représentatif : sept catégories fonctionnelles ont été définies sur un ensemble de critères (touchant aux caractéristiques physiques des sites et de leur environnement, à l’iconographie et aux activités pratiquées sur place), et traduits en comportements sociaux. Les plus nombreux — sites rituels occasionnels (casual ritual sites) et rochers ornés jalonnant les déplacements au sein du paysage (landmarks ou waymarks) — attestent d’une production et d’une utilisation d’images rupestres dans un cadre quotidien engageant l’intégralité du groupe. Leur fonction est toutefois tout autant séculière que rituelle. D’autres sites ornés seraient avant tout dédiés à la fonction « religieuse », dans deux cadres sociaux clairement distincts : les sites d’agrégation mettant en jeu l’organisation sociale étendue (inter-groupes), et les « sanctuaires-ermitages » impliquant un public restreint dans une démarche contemplative et introspective, peut-être pour les initiations. La distribution des diverses catégories fonctionnelles de sites ornés révèle un paysage social complexe à l’intérieur d’une même entité écologique, composé de différents espaces ouverts et entremêlés pour la plupart, segmenté dans le cas des sanctuaires-ermitages. Ces « discontinuités symboliques synchroniques » dans l’espace livrent un nouvel éclairage sur la flexibilité spatio-temporelle des formations sociales de ces populations.
Changer ou tenir ?
Identités et relations sociales ne sont pas des états de fait mais des constructions culturelles en perpétuelle renégociation, et la production graphique en constitue l’un des agents, pour les individus et pour les groupes sociaux. Elle ne s’abstrait pas des mêmes enjeux (reproduction, réajustement, rupture). Performative et interactive, elle est ainsi une forme d’action sociale, chargée d’une force d’action en retour sur la société. Ainsi, bien que les populations de chasseurs- collecteurs soient vues comme “traditionnelles” et donc conservatrices, en particulier dans leur art rupestre en raison de sa portée collective et le plus souvent spirituelle et rituelle, différentes dynamiques ont été mises en lumière dans leurs iconographies : conservatisme, emprunts et remplacements thématiques, créolisations, ruptures. Ces variations diachroniques peuvent intervenir en réaction à une autre société, mais sont aussi documentées en l’absence d’un tel stimulus.
À l’instar de l’iconographie, les paysages d’art rupestre ne sont pas immuables : actifs car intégrés à l’identité d’être dans le monde, ils sont discutés et soumis à variations dans le temps. Les réseaux de sites ornés — implantation, nombre, relations spatio-temporelles dans le cycle de mobilité — peuvent donc évoluer dans la diachronie pour différents motifs : contrainte physique d’accès aux lieux, reconfiguration de la mobilité, restructuration de l’organisation sociale, transformation du système ontologique. La balance entre facteurs externes — changements environnementaux, arrivée d’une population étrangère — et dynamiques internes dans l’impulsion de ces évolutions sur la longue durée a été au coeur de nombreuses recherches abordant l’angle théorique de l’Information Exchange Theory.
Ces questionnements sont au coeur de recherches en cours sur la variabilité socio-culturelle des populations de chasseurs-collecteurs du massif des Matobo (Zimbabwe) entre fin Pléistocène et Holocène récent, notamment axées sur les dynamiques de la production rupestre (programme MATOBART : dir. Bourdier 2019). Plusieurs milliers de parois rocheuses — rochers à l’air libre, petits et grands abris-sous-roche — ont été investies à travers tout le massif, montrant une grande diversité en termes de topographie et de morphologie des lieux, d’iconographies et de contextes archéologiques. Un volet de ce programme interdisciplinaire interroge la stabilité des paysages d’art rupestre, pistant les « discontinuités symboliques diachroniques » dans les réseaux de sites ornés afin de les confronter aux données issues de la sphère environnementale et de la culture matérielle. Dans quelle mesure les changements écologiques ont-ils impacté l’expérience que les sociétés avaient de leur espace ? En tant que constructions socio-culturelles mettant en jeu des éléments de l’environnement, paysages culturels et culture matérielle co-évoluent-ils ?
Un premier modèle d’occupation du massif des Matobo et de ses dynamiques sur le temps long a été élaboré par N. Walker, incluant quelques éléments sur l’art rupestre : intensité, nombre et fonction socio-culturelle des sites ornés. Il met en exergue de fortes dynamiques démographiques au cours des millénaires, en réponse directe aux fluctuations climatiques selon l’auteur. Il reconnaît de nombreux changements dans l’organisation sociale (taille et composition du groupe) et la structuration territoriale des populations (surface, saisonnalité, type de mobilité, modalité de la mobilité résidentielle). Ces transformations seraient en partie dictées par les ressources alimentaires disponibles qui dépendraient elles-mêmes des conditions climatiques et de la démographie ; elles relèveraient toutefois également de mécanismes internes comme l’atteste la diversité des comportements développés. La production graphique serait l’une des réponses comportementales collectives apportées à un stress social, notamment à l’arrivée de populations allochtones. Un épisode de forte démographie à l’Holocène ancien ferait basculer la fonction de l’art rupestre sur des enjeux communautaires inter-groupes (guérisons, initiations). Liées à une transformation des réseaux sociaux et de la flexibilité sociale (mise en place du système de dispersion/agrégation), trois catégories de sites ornés coexisteraient depuis la fin de l’Holocène moyen : site spécialisé (specific art sites), site d’habitat (small living sites) et méga-site (collective ceremonial centres).
Face à ce modèle, la stabilité des paysages d’art rupestre se doit d’être interrogée ; ou plutôt les moments, rythmes, modalités et facteurs de changement se doivent d’être examinés et précisés. En effet, plusieurs « styles » graphiques coexistent au sein des ensembles rupestres des Matobo, dont certains manifestement inscrits dans la diachronie au regard des stratigraphiques pariétales. Or, tous ne se retrouvent pas de manière concomitante dans les sites. La diversification fonctionnelle des sites d’art rupestre devrait aller de pair avec des distributions différentes entre les catégories de sites, en termes de nombre et d’implantation (plus de contraintes matérielles pour les méga-sites). Le paysage d’art rupestre pourrait ainsi s’élargir au sein du massif, avec des choix d’implantation peut-être originaux par rapport aux paysages antérieurs de l’art rupestre. À la lecture du modèle de Walker, on serait par ailleurs en mesure de s’attendre à d’autres évolutions : augmentation du nombre de sites et distribution plus large au sein du massif au cours des épisodes de contraction territoriale en raison d’une forte démographie, d’une occupation tout au long de l’année, d’une zone d’exploitation résidentielle resserrée ; sites moins nombreux et plus localisés dans l’espace quand les territoires s’étendent en-dehors du massif (occupation saisonnière, zone d’exploitation résidentielle étendue) ou à cause d’une faible démographie.
L’enjeu sera d’analyser les discontinuités ainsi que les permanences. Les lieux investis sur le temps long par des générations successives questionnent la mémoire sociale et l’attachement aux lieux et aux paysages, et explorent plus spécifiquement le pouvoir récursif de l’art rupestre. Dans quelle mesure l’iconographie s’inspire-t-elle du passé, l’incorpore-t-elle, l’utilise-t-elle ? À travers la récursivité des lieux et en leur sein des images, une dimension ontologique supplémentaire est désormais effleurée : celle du rapport des populations de chasseurs-collecteurs préhistoriques à leur passé.
Conclusion
Les sites d’art rupestre ont contribué à l’expérience de l’environnement, de l’espace et plus généralement du monde des nomades chasseurs-collecteurs du passé, et participé à la définition de leurs paysages socio-culturels. Les paysages d’art rupestre ont tenu un rôle actif dans les stratégies d’occupation des espaces, de mobilité et de flexibilité sociale. Ils ouvrent une vision anthropologique plus dynamique sur la production et l’utilisation des images au sein de ces sociétés dont ils oeuvrent à éclairer la diversité, le dynamisme et la créativité.
Entrées d’index
Mots-clés : art rupestre, paysage, Information Exchange Theory, Afrique, Matobo, Brandberg
Keywords : rock art, landscape, Information Exchange Theory, Africa, Matobo, Brandberg
Texte intégral
Introduction
1Human populations, and nomadic ones in particular, live their environment; they perceive, occupy, and interact with it according to intentions and modalities that are specific to their ontological worldview, ultimately transforming this environment into a landscape. From an anthropological point of view, a landscape can be defined as a cultural construct insofar as it constitutes “a set of conventional and normative understandings through which people construct and make sense of their cultural world” (Tilley 1996: 161). These constructions are at once the expression of a physical reality anchored within a given material space; particular natural conditions (biological or mineral resources) that can be exploited (for alimentation, raw materials, or for shelter); and also an idealized perception (conception of and relation to the world outside of ego). Born of this dialog between an environmental context, its perception and the resulting behavioural responses, landscapes are culturally and socially constituted, as well as historically situated (Thornton 1980; David & Lourandos 1999; David & Wilson 1999).
2Different metaphysical spaces and/or zones of physical exploitation structure landscapes at various scales and diverse geometries. They are punctuated by places, portions of structuring spaces whose functions (technical, alimentary, ritual, historical, etc.) and modes of occupation (access, time, length) are culturally and socially significant. The resonance of the environment among human societies gives certain places a specific affective dimension, notably (though not exclusively) associated with the spiritual domain (Rapoport 1975). This network of known and defined places actively participate in the constitution of both collective and individual identities (Sack 1980; Tilley 1996). Through a double dialectic, populations self-identify and are identified by the “other” in the ways that they are within the space they occupy both physically and mentally (territories). Among the !Kung1 hunter-gatherers of southern Africa, each band is linked to places with essential resources (n!ore), such as sources of water, and the rights access to these spaces are inalienable and transmitted through the generations, even if temporary access can be granted to others through the navigation of other social relationships, such as those mediated by the reciprocal non-equivalent delayed exchange network known as hxaro (Marshall 1976; Wiessner 1982). More than a form of control of a space that permits groups and individuals to benefit from the dispersed resources of an inhospitable environment, this 'right' should be understood as a collection of relationships - and therefore obligations - linking people to places (Smith 1994). The link between identity and landscape is even stronger among the aboriginal populations of Australia, where socio-economic organization - including kinship - and the metaphysical sphere are linked to highly significant places that were created by the actions of mythical ancestors during the Dreaming. These links create numerous obligations, notably ritual ones (Layton 1992; among numerous others).
3Mobility actively contributes to the construction of identity within and between places: it permits experience, both in terms of perception and apprenticeship, knowledge and cultural appropriation, both affective and social (Bradley 1997; David & Wilson 1999). Socially regulated (time, periodicity, route, direction), this spatiotemporal cycle of territorial occupation, simply by its recursive nature, anchors, develops, and renews links between societies, landscapes, and places in a continuous act of co-construction.
4Rock art sites can be a major component in the construction of the physical, mental, emotional, and affective experiences that constitute landscapes. They possess three particularities: (1) they are “marks” (intentional traces) left on the environment that use an (2) immovable and (3) perennial medium, ultimately fixating the location of these marks in space and inscribing them in the long-term. These three properties give rock art sites a remarkable dimension among nomadic hunter-gatherers as few other actions or creations of such populations are intended to leave permanent and immobile material traces in the environment. Rock art “inscribes” the landscape (David & Wilson 2002), demonstrating its high significance. It reinforces the relationship between populations and their lived and thought space, all the while contributing to the organization and anchoring of socio-cultural practices within space, ultimately punctuating this space with multiple physical and metaphysical markers.
5In archaeology, landscapes defined by rock art sites provide a privileged access point for apprehending the territorial structure2 of populations. The repartition of decorated sites and their interrelationships are an aspect of mobility strategies and human occupation within landscapes that is potentially complementary to the techno-economic sphere. The ultimate transdisciplinary goal is to understand these “landscape engagements that are rich in cultural meaning” (David & Hamasher 2018), another facet in our quest to reconstitute the cultures and societies of the past. This line of research has been in full development in the archaeology of peoples without writing across the globe since the 1990s, with numerous works broaching notably the choice of places, whether via investigations of physical environmental characteristics and the use of spatial analysis toolsets or by the use of phenomenological and sensory approaches to landscapes (Chippindale & Nash 2004; Tilley 1996; Bradley 1997; among others). This contribution aims to continue along this path by reflecting upon the heuristic potential of rock art landscapes for the archaeology of southern Africa's hunter-gatherers using two examples. The first will expose a model for the spatial organization of rock art sites advanced by T. Lenssen-Erz in Brandberg/Daureb (Namibia), which elucidates certain aspects pertaining to the perception of this landscape and its socio-cultural organization by groups during the Late Holocene. The second, a product of current analyses, investigates the diachronic stability of rock art landscapes on the long-term, between the end of the Pleistocene and the Late Holocene in the Matobo hills (Zimbabwe).
Rock art landscapes as multiple facets of past social landscapes
6Rupestrian graphic production is a component of the symbolic sphere of human societies, notably being one of likely many outward expressions of their spirituality. Ethnographic and ethnohistoric sources on hunter-gatherer populations indicate that the creation and use of this iconography are most often involved in ritual practices of diverse natures: rain; animal, plant, and human reproduction; the control of game movements and hunting magic; healing; divination; reactivation of the power of certain divinities, spirits, ancestors, or mythical beings; male and female initiation rites (Vinnicombe 1976; Lewis-Williams 1981; Morphy 1991; Layton 1992; Lim 2010; Whitley 2010; McDonald 2013; McGranaghan & Challis 2016). In fact, several ritual contexts can be cumulated within one society. Sources also reveal that the site chosen is often highly significant, actively participating in the symbolism of images of the efficiency of the conducted ritual itself. The mythical landscape associated with the Dreaming of Australian Aborigines is a well-known example: rock art sites are prominent elements within the mental map of the world's cosmogony that is partly materialized in the environment, many of which are attached to specific episodes of the lives of mythical beings or ancestors, their creation, or elements and phenomena that interact with the real world. Engravings or paintings are often said to be the result of mythical beings, notably in that they are the traces or images left behind when said entities entered or left the world through the rock surface (Morphy 1991; Layton 1992; Chaloupka 1993; McDonald 2013). Certain rock art sites of the Karoo people of South Africa are seen as mythical places where the bodies of primordial beings were transformed into elements of the natural environment (Deacon 1988). Rock art sites can also be places that are invested with a particular power linked with a supernatural dimension, to which the rock surface provides an opening, such as in described shamanic contexts (Lewis-Williams 1981; Whitley 2010). However, rupestrian graphic production cannot be exclusively tied to the spiritual and ritual realms, as some can reveal secular (economic or pedagogic) or even anecdotal motivations (Layton 1992; Guenther 1994). Rupestrian iconography also reflects diverse sociocultural motivations and intervenes in multiple and varied semantic contexts.
7All rock art sites cannot, therefore, be considered at first sight as the many links that constitute a sacred or ritual landscape. Faced with the difficulties involved in identifying the motivation behind a symbolic work and in historicizing emic knowledge and transposing it in the long-term, numerous studies have instead chosen to broach the larger social dimensions prehistoric of rock art, such as, for example, its role in the manufacture, transmission, and regulation of information (Conkey 1980, 2010; Lenssen-Erz 1994; Solomon 1994; McDonald & Veth 2012; among others). As a medium and mediator for information exchange, graphic representation actively intervenes in the negotiation of individual and collective social identities, as well as in intra- and inter-group relations (Information Exchange Theory: Wobst 1977; Wiessner 1984). The role of rock art is therefore linked to social context (individuals present and associated activities) and is correlated with place and time. From this anthropological point of view, rock art sites become windows into the social dynamics of populations within space and inscribed within their environment (Conkey 1980, 2010; McDonald & Veth 2012; Bourdier 2013; McDonald 2015, 2017; Bourdier et al. 2017). Far from being arbitrary, their insertion is the result of decision-making processes that associate not only conceptual/spiritual/symbolic factors previously discussed, but also material necessities (access, size, protection, water, and food). The dialog between “ideology” and “practice” (Lenssen-Erz 2004) that occurs as humans interact with their environment makes rock art landscapes eminently cultural (and therefore spatially and temporally situated) and socially mutable.
8The work of T. Lenssen-Erz and the team from the University of Cologne (Lenssen-Erz 2001, 2004, 2008) has therefore put the spotlight on all the complexity that constituted the socio-cultural landscape of the Brandberg/Daureb intrusion (Namibia) occupied by hunter-gatherers during the Late Holocene (4000-2000 non cal. BP). With an altitude of 2200 m, this roughly 570 square kilometre inselberg sitting on the doorstep of the Namib Desert was a living space that was privileged by hunter-gatherer populations of the Later Stone Age because of its rich ecosystem positioned within one of the most arid regions of the continent. This area houses one of the densest concentrations of hunter-gatherer rock art in southern Africa, with roughly 1000 sites and 50 000 images, all documented by the incredible work of inventory and tracing conducted by H. Pager and his assistants. Eminently figurative, the iconography mixes humans, who form the majority, with animals, mostly large herbivores within which diverse species of antelope, notably the springbok, are also documented. The refined technical nature (frequently polychromatic, very fine lines) and the “naturalism” of the figures (degree and detail of their dynamism) are responsible for the reputation of these rupestrian paintings. Sites are diverse in their morphology (boulders, shelters) and topography, and are distributed all across the intrusion with, however, drastically varying graphic densities, with most sites only containing and handful of representation while a few house hundreds (figure 15.1).
9A model was developed in order to interpret the sociocultural function of rock art sites in the Brandberg/Daureb using a sample of 327 sites dispersed over 125 km². The sample was representative of the topographic diversity of the intrusion and numerous criteria (physical characteristics of sites and their environment, iconography, activities practiced on site) were taken into account in order to translate the sites into social behaviours (Lenssen-Erz 2004, 2008). Presentation (public/private), frequentation (minimum/maximum), ritual activity (minimum/maximum), primary motivation (materialist/metaphysical3) and primary function (secular/religious) all contribute in determining the seven functional categories for rock art sites. (table 15.1). The most numerous are the casual ritual sites that would have been used by very mobile small groups in response to sudden needs (healing, rain). Similarly, decorated boulders mark movements within a landscape (landmarks or waymarks) and document the production and use of rupestrian images by all members of the group as a part of daily life, with their function being just as secular as they are ritual.
Functional categories | Qualities (social behaviours) | Environment, topography, site morphology | Imagery | Archaeological context (activities) |
Short-term living site | Secular function | Few | ||
Private presentation | ||||
Materialistic motivation | ||||
Minimum ritual activity | ||||
Minimum frequentation | ||||
Long-term living site | Secular function | Few | Density and diversity of artefacts | |
Private presentation | ||||
Materialistic motivation | ||||
Minimum ritual activity | ||||
Maximum frequentation | ||||
Landmark/waymark site | Secular function | Along natural travel routes | Few | Few |
Public presentation | Nor / very poor accommodation | |||
Materialistic motivation | No further natural ‘infrastructure’ | |||
Minimum ritual activity | Near to conspicuous landscape features | |||
Minimum frequentation | ||||
Aggregation camp | Secular function | Large living sites | Very high number of paintings | High density and diversity of artefacts |
Public presentation | Open areas in the immediate surroundings | Strong emphasis on women | High intensity of usage | |
Idealistic motivation | Good natural ‘infrastructure’ | |||
Maximum ritual activity | Many small satellite-sites in the near vicinity | |||
Maximum frequentation | ||||
Sanctuary-hermitage | Religious function | Special position (panoramic view) | Few paintings | Few |
Private presentation | Isolated location | Extraordinary paintings | Little intensity of usage | |
Idealistic motivation | Partly difficult access | |||
Maximum ritual activity | Peculiar rock formation | |||
Minimum frequentation | Disadvantageous natural ‘infrastructure’ | |||
Reduced roofed area | ||||
No open area outside | ||||
Deliberate ritual site | Religious function | Medium/high number of paintings | Low intensity of usage | |
Private presentation | ||||
Idealistic motivation | ||||
Maximum ritual activity | ||||
Maximum frequentation | ||||
Casual ritual site | Religious function | Rather small sites | Medium/high number of paintings | Low intensity of usage |
Public presentation | Mediocre accommodation facilities | Rather complex paintings | ||
Materialistic motivation | Mediocre access to natural ‘infrastructure’ | |||
Minimum ritual activity | ||||
Minimum frequentation |
10The area also has decorated sites that would be, above all else, dedicated towards a “religious” function albeit in two clearly distinct social contexts: aggregation camps and sanctuaries-hermitages. The role of the first would be to reaffirm, maintain, and develop social links and identities between affiliated groups: graphic representations would be situated at the heart of strong interactions that bring into play the extended social organization (between groups), accompanied by rituals and, more largely, a collection of highly standardized social behaviours. In regards to the preeminent role given to the representation of women in these sites, Lenssen-Erz (2008) put forth the hypothesis that women were the “custodians of the community” and the “masters of social codes”. At the opposite end of the spectrum, sanctuary-hermitage sites would have been frequented by a restricted public with a more contemplative and introspective approach, possibly having served in initiations. This would explain their specific locations, at the highest altitudinal level, where the extended view would permit individuals to embrace the entirety of the landscape in order to learn to read it, its history, and its significance.
11This research reveals a mosaic landscape of rock art in the Brandberg/Daureb. The distribution of diverse functional categories of decorated sites brings to light a socially complex landscape that sits within a single ecological entity, composed of different social spaces - between group, communal and within group, restricted to certain categories of individuals as defined by their age and gender (initiations) - open and mixed for most, yet segmented and restricted in the case of sanctuary-hermitages. These “synchronically symbolic discontinuities” (Hernández Llosas 2012) in space shed new light on the spatiotemporal flexibility of social formations and the structure of sociocultural networks of these hunter-gatherer populations of the Late Holocene.
Change or remain? Rock art landscapes, social dynamics, and the long-term in the Matobo Hills (Zimbabwe)
12Graphic production is not simply a passive medium through which messages are transmitted. It possesses its own capacity for action and reaction, both on the author of the message and those who receive it by way of engaging them in an oscillating discourse between acceptance and refusal of the conveyed message through a gradation of revisions and modifications. Both performative and interactive, graphic production is a form of social action that has an active influence on society: it can contribute just as much to transformation as it can to conservation (Lewis-Williams & Dowson 1994; Gell 1998; Conkey 2010). Identities and social relationships aren't stable facts, but rather cultural constructions in perpetual renegotiation via processes of reproduction, readjustment, and rupture. Graphic production is an active agent is such renegotiations, for both individuals and social groups, and is therefore not freed from the same issues or resilience or change.
13While nomadic hunter-gatherer populations are often seen as 'traditional', and therefore conservative (Whitley 2010), especially in their rock art given its collective and often spiritual and ritual scope, their variable iconographies have shown distinct dynamics. In particular, numerous studies regarding 'contact' with other populations —nomadic pastoralists, agriculturalists and herders, colonizers— reveal diverse processes: conservatism as an expression of identity, spiritual resistance and the rejection of external influence (Layton 1992); borrowings and thematic replacements within the context novel of socio-economic cooperation that brings into play the spiritual domain (Dowson 1994; Hall 1994; Yates et al. 1994; Eastwood et al. 2010); even creolization within processes of novel mixed identities (Challis 2014). Within these contexts, graphic production is the instrument to fight against, legitimize, consolidate or even reinforce a new sociocultural order. Ruptures or discontinuities in graphic traditions are still documented even in the absence of cohabitation between distinct societies, notably in Australia, which calls into question the nature of the varying factors in play, be they economic, ecological, or social (Layton 1992; McDonald 2015, 2017; David et al. 2019; among others). As a matter of fact, numerous ethnographic studies have shown the fluidity, flexibility, significant capacity for adaptation, and absence of normalization of beliefs and rituals among certain hunter-gatherer populations (Guenther 1994).
14As seen with iconography, rock art landscapes, interpreted as sociocultural constructions, are not immutable: the distribution of decorated sites and their interrelationships are also subjected to a permanent tug-of-war between continuity and change. Interactions between human groups and their landscapes are not fixed but are rather active as they are integrated into identity construction within the material world. These interactions are therefore discussed, negotiated, and subject to temporal variations (David & Lourandos 1999). During the natural course of life and mobility the Sandawe of Tanzania invest novel places as they decorate them for rain rituals, and abandon old ritual sites (Lim 2010). In Queensland (Australia), the current cultural landscape of the Ngarrabullgan Mountain, within its ontological context of the Dreaming (which includes painted sites), was constructed roughly 600 years ago, demonstrating a profound renewal of the conceptual and physical relationships between the world and the Aboriginal populations inhabiting this region (David & Wilson 1999). Just as with other dimensions of landscapes, networks of decorated sites —localization, number, and spatiotemporal relations within a mobility cycle— can diachronically evolve for different reasons: limits to physical access, evolution of the mobility cycle or strategy, restructuring of social organization, or transformation of the underlying ontological system. The balance between external factors - environmental change, arrival of external populations- and internal dynamics in the initiation of these long-term evolutionary processes were at the heart of numerous studies investigating hunter-gatherer rupestrian art using the theoretical framework provided by Information Exchange Theory. The first models provided within this framework were heavily influenced by varying degrees of ecological determinism (Gamble 1982; Jochim 1983), but these eventually gave way to more complicated images. Research conducted in the Western Desert and on the Murujuga Peninsula in Western Australia revealed diverse social responses to environmental change, by way of socio-economic and territorial restructuring. The production of rock art filled distinct social roles between these different Aboriginal populations, signalled by differences in iconography and the organization of networks of decorated sites (McDonald 2015, 2017).
15The questions are at the heart of the research that is currently being conducted on the diachronic dynamics of rupestrian graphic production in the Matobo of Zimbabwe (the MATOBART program: Bourdier 2019). This project explores the sociocultural variability of hunter-gatherer populations of the region in the long term (end of the Pleistocene through to the Late Holocene), and one of the parts of this interdisciplinary program investigates, in particular, the structure of territories through the interpretive, and privileged, lens of rock art landscapes. Through this study of long-term human-place relations, our goal is, on the one hand, to track “diachronically symbolic discontinuities” (Hernández Llosas 2012) in the networks of decorated sites, and, on the other hand, confront these results with data from the environmental and material culture domains. To what degree did ecological changes impact how societies experienced their “space”? Do material culture and cultural landscapes, both being cultural constructions, coevolve together?
16Covering over 2000 square kilometres, the Matobo hills, much like the Brandberg/Daureb region, are among the principal areas of rupestrian art in Southern Africa, with an estimated 3000 painted sites, the majority of which have been attributed to the “fine-line painting” artistic tradition of hunter-gatherers (Walker 1996; Nhamo & Bourdier 2019). The area contains a national park that has been classified by UNESCO as a “cultural landscape” because of its exceptional rupestrian heritage, in addition to the spiritual and ritual links that local and regional communities continue to maintain with various places - be they anthropic in nature or not - in the landscape (Mwali religion). As seen in the Brandberg/Daureb region, the imagery highlights humans engaged in multiple actions (hunting lines, groups walking, campsite scenes, dance scenes) associated with a wide variety of animals where large herbivores are dominant (antelopes, giraffes, elephants). While more discreet, references to plant life are recurrent, most notably via the representation of trees and shrubs, exceptionally depicted in the iconography of hunter-gatherers across the world. These paintings are particularly famous for their “naturalism”, their use of very fine lines, as well as their frequent polychromatic nature, using a broad spectrum of colours (figure 15.2).
17Multiple locales across the area have been decorated, ranging from freestanding boulders to large or small rock shelters, demonstrating the large topographic and morphological diversity of places, in addition to diversity in iconographies and archaeological contexts (figure 15.3). Both the density and composition of representations vary in different sites, within which distinct technical and formal treatments sometimes meet, treatments whose characterization and chronological variability count among the major issues the MATOBART program seeks to resolve. Grounds provide evidence for diversity, from an absence of traces to a whole spectrum of deposits, sometimes several metres thick, that are formed by the succession of numerous layers that are variably thick and/or dense and contain the vestiges of past activities attributable to the alimentary, technical, and sometimes symbolic (personal ornaments) sphere.
18After numerous archaeological interventions throughout the first half of the 20th century, an impressive bit of research was conducted by N. Walker during the 1970s and the 1980s, mixing archaeological test pits and the restudy of old collections. This work resulted in the proposition of a model for the dynamic long-term occupation of the hills, between the end of the Pleistocene (roughly 13 000 BP) to the Late Holocene (roughly 1 500 BP), when the archaeological signature of hunter-gatherers in the area disappears (Walker 1995). This model proposes links between demography, territorial structure, and environmental conditions by using data from the industrial component (lithic and osseous industries), data regarding subsistence strategies (fauna and flora exploited), as well as number sites and the intensity of their occupation (extrapolated from the absolute volume of remains excavated) for each chronological phase (table 15.2). It includes a few elements from the symbolic sphere, mostly regarding the intensity of graphic representation (extrapolated from the volume of pigments within archaeological deposits) as well as the quantity and sociocultural function of decorated sites (Walker 1994).
Occupation phases | Climatic conditions | Diet focus | Demography | Settlement model | Ritual activity (including rock art) | Rock art sites |
Phase 1 (˃ 22 ky BP) | Cold and dry | No rock art | ||||
Late Pleistocene (MSA) | ||||||
Phase 2 (22-13 ky BP) | Pick of aridity | No rock art | ||||
Late Pleistocene (LSA) | ||||||
Phase 3 (13- 9,5 ky BP) | Cool and wet | |||||
Pleistocene/Holocene transition | ||||||
Stage 3a (13-10.5 ky BP) | Big game hunting | Low population | Autumn presence | Low production | Big living sites | |
Very large residential groups | ||||||
High mobility | ||||||
Very large home range | ||||||
Single central residential camp | ||||||
Stage 3b (10,5-9.5 ky BP) | Big and small game hunting | Increasing population | Autumn presence | High production | Big living sites | |
Increasing fruit exploitation | Large residential groups | |||||
Large home range | ||||||
High mobility | ||||||
Several residential camps | ||||||
Phase 4 (9,5-7 ky BP) | Hot and damp | |||||
Early Holocene | ||||||
Stage 4a (9.5-8.5 ky BP) | Intense fruit exploitation | Highest population | Summer/Autumn or Year-round residency | High to low production | Big living sites | |
Small game hunting | Large residential groups | More public inspired painting | ||||
Compacted home range | Shift from big game depictions to humans | |||||
Low mobility | Shift from manipulative to social ritualistic function (curing shamanic ceremonies, initiations) | |||||
Spaced residential camps | ||||||
Stage 4b (8.5-7 ky BP) | Big and small game hunting | Decreasing population | Year-round residency | Low to high production | Specific art sites | |
Decreasing fruit exploitation | Large residential groups | |||||
Random residential camps | ||||||
Large home range | ||||||
Competition / social boundary display | ||||||
Phase 5 (7-4,5 ky BP) | Hot and damp | |||||
Mid Holocene | Aridification | |||||
Stage 5a (7-6 ky BP) | Big and small game hunting | Low population | Year-round residency | Low production | Specific art sites | |
Decreasing fruit exploitation | Large residential groups | |||||
Random residential camps | ||||||
Large home range | ||||||
Stage 5b (6-4.5 ky BP) | Small game hunting | Low population | Year-round residency | Low production | Specific art sites | |
Small residential groups | + small living sites | |||||
Large home range | + collective ceremonial centres | |||||
Low mobility and isolation | ||||||
Periodic short aggregations | ||||||
Phase 6 (4,5-1.5 ky BP) | Present | |||||
Late Holocene | ||||||
Stage 6a (4.5-2.2 ky BP) | Big game hunting | Growing population | Year-round residency | Low production | Specific art sites | |
Incoming population | Aggregation/dispersal | + small living sites | ||||
Large home range | + collective ceremonial centres | |||||
High mobility | ||||||
Stage 6b (2.2-1.5 ky BP) | Domestic cattle | High population | Year-round residency | High production? | Specific art sites | |
Big game hunting | Aggregation/dispersal | + small living sites | ||||
Large home range | + collective ceremonial centres | |||||
High mobility | ||||||
Pastoralism? |
19This model emphasizes strong demographic dynamics over several millennia, interpreted by Walker as being the response to climatic fluctuations. Walker recognizes numerous changes in the social (size and composition of groups) and territorial (surface, seasonality, type of mobility, means of residential mobility) structuring. These transformations would have been dictated in part by the availability of different resources, which would have in turn been dependent on both climatic and demographic conditions; though these transformations would also reveal internal mechanisms, as demonstrated by the diversity of developed behaviours. Graphic production would have been one of the collective behavioural responses to social stress, notably provoked by the arrival of non-native populations. It is, moreover, exactly such a demographic episode that Walker theorizes as being the catalyst for the beginnings of a shamanic type system in the Early Holocene, accompanied therefore by a shift in the function of rock art in regards to community issues (healing and initiation rites). While the shamanic sociocultural framework would remain, the social contexts for such activities would diversify over time, with graphic representation being initially under the purview of specific art sites, but eventually moving into small living sites and collective ceremonial centres.
20Certain aspects of the model require further discussions and reevaluation (notably the calculation of site occupation intensity and graphic production), but the stability of rock art landscapes, given this long-term model of dynamic territorial structure of the Matobo hills, requires some much-needed investigation. Or rather, the moments, rythms, means, and factors of change need to be specified and examined. It is true that several graphic “styles” coexist within the rupestrian assemblages of the Matobo hills, and certain styles demonstrate a diachronic aspect when they are considered in terms of parietal stratigraphic superpositions. However, not all of them can be found concomitantly in the sites, demonstrating variation in the networks of painted sites. Walker's model already proposes a first hypothesis pertaining to a discontinuity with the diversification of rock art sites at the end of the Middle Holocene (stage 5b) that he links to a transformation in social networks and a degree of social flexibility that accompanies the implementation of a dispersal and aggregation system. Following this logic, the number of sites from each of his three categories should vary, just like their distribution, with a high number of specific art sites and a low number of collective ceremonial centres. The collective ceremonial centres, which would correspond with aggregation sites, would also have specific needs in terms of size and sufficient water and food in proximity in order to sustain a large number of people, and let us not forget that the Matobo hills are found within a semi-arid environment where water sources are highly dispersed and often ephemeral. This functional diversification of rock art sites would therefore be concomitant with a greater distribution within the hills and a greater diversity of choices regarding location, perhaps even novel choices (stages 5b and 6a), relative to the preceding rock art landscape. In reading Walker's model, we would also expect other systemic evolutions:
- - an augmentation in the number of sites and a wider distribution across the hills during episodes of territorial reduction as it relates to high demographic pressure, year-long occupation, and restricted zones of residential exploitation (stages 4a and 4b);
- - a reduction in the number of sites and a reduced spatial distribution of said sites when territories expand beyond the limits of the hills (seasonal occupation, extended zone of residential exploitation: stages 3a and 3b) or because of reduced demographic pressure (stage a).
21Is the “remarkable diversity of behaviour” (Walker 1995: 258) that is apparently observed among the nomadic hunter-gatherer populations of the Matobo hills echoed in the rock art landscapes? The stakes will now revolve around the analysis of not only the observed discontinuities, but also that which is permanent: the locations that demonstrate a repeated investment by successive generations over the long term. In the face of such transformed social organizations, spatiotemporally displaced occupation cycles, and restructured social networks, an observed longevity in the use of specific rock art sites would evoke special and highly significant locales. This opens up lines of questioning regarding social memory and attachments to places and landscapes (Tilley 1996), and more specifically the dialectic between past and present in graphic creation. To what degree are the images produced inspired by the past, and do they incorporate or use past images? As underlined by Morphy, rupestrian art has a recursive power: “Rock art provides a reservoir of images for succeeding generations who can not only view and interpret the record, but also use it as a source of information that influences their present practice” (Morphy 2012: 294). In the Drakensberg (South Africa), the incorporation of ancient images into new ones, or reinterpretations in which certain actors (themes) are transformed using similar narratives, have been interpreted as demonstrating the will of the authors to subscribe to a symbolic, spiritual, and ritual continuity (Lewis-Williams 1981; Hall 1994). Through the recursive nature of places and the images within them, a novel ontological dimension is revealed; that of the relationships of societies with their past, and in this instance the relationship between prehistoric hunter-gatherer populations and their past.
Conclusion
22Just as with sites whose functions revolve around the technical or the alimentary, or as with living sites, rock art sites contributed to hunter-gatherer experiences of the environment, space, and the world more generally, and participated in the construction of their socio-cultural landscapes. Forming an essential link between the world and the symbolic, spiritual, and ritual spheres, rock art landscapes had an active role in territorial organization, in strategies for the occupation of space, mobility, and social flexibility. By interrogating the spatial, temporal, and functional complementarity of places, we open the door on a more dynamic anthropological vision of the production and use of images within these societies. Research conducted in the Brandberg/Daureb demonstrates this complexity, which touches upon the multiple social spaces that intervene during the construction and embodiment of individual and collective social identities. The research underway in the Matobo hills, exploring the diachronic variability of rock art landscapes, will inform us on the history of behavioural responses to environmental, economic, and social changes. Both projects aim to elucidate the diversity, the dynamism, and the socio-cultural creativity of past nomadic hunter-gatherers.
Acknowledgements
23The author wishes to acknowledge the French Ministry of Europe and Foreign Affairs (MEAE), the Institut Universitaire de France, the French Embassy in Zimbabwe, The French Institute in South Africa (IFAS), the CNRS, and the UMR 5608 TRACES for their financial and institutional support to the MATOBART research program. The author wants to thank the four reviewers for their accurate reading and relevant advices. The author is especially grateful to T. Lenssen-Erz and E. Fäder for the pictures, and their wholehearted support.
Notes de bas de page
1 Group that belongs to the San speaking hunter-gatherer populations.
2 Definition and organization of frequented spaces, i.e. their spatiotemporal interrelationships and associated social dynamics (flexibility in social formations).
3 A rock art site that was guided by a materialist motivation is one that was initially chosen for practical reasons associated with the imperatives of daily life (protection, food, and water in proximity), with the act of painting happening in a non-planned and circumstantial fashion (Lenssen-Erz 2004: 140-141). A metaphysical motivation (idealistic motivation) implies that the site was chosen to be painted because of its significance and ritual and religious power, all intrinsically identifiable at the site. It is therefore not the more or less circumstantial production of paintings that can attribute such a dimension after the fact (Lenssen-Erz 2004: 144).
Auteur
Archaeologist, Associate professor (Senior Lecturer) at the Toulouse Jean-Jaurès University, Junior member of the Institut Universitaire de France (IUF), Honorary Research Fellow Rock Art Research Institute, University of the Witwatersrand - UMR 5608 – TRACES, Maison de la Recherche - Bât 26, 5, allées Antonio MACHADO, 31058 Toulouse Cedex 9 (France)
Le texte seul est utilisable sous licence Licence OpenEdition Books. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.
Not Just a Corridor
Human occupation of the Nile Valley and neighbouring regions between 75,000 and 15,000 years ago
Alice Leplongeon, Mae Goder-Goldberger et David Pleurdeau (dir.)
2020
La langue des bois
l’appropriation de la nature entre remords et mauvaise foi
Sergio Dalla Bernardina
2020
Living and working with giants
a multispecies ethnography of the Khamtis and elephants in Northeast India
Nicolas Lainé
2020
Nomad lives
From Prehistoric Times to the Present Day
Aline Averbouh, Nejma Goutas et Sophie Méry (dir.)
2021
L’art des ghostnets
Approche anthropologique et esthétique des filets-fantômes
Géraldine Le Roux
2022
Elusive Partners
Contemporary Anthropological Perspectives on Marine Species
Alix Levain, Hélène Artaud, Émilie Mariat-Roy et al.
2023
Des voies de l’ombre
Quand les chauves-souris sèment le trouble
Frédéric Laugrand et Antoine Laugrand
2023