Ce travail de thèse présente une nouvelle lecture technologique des industries lithiques qui documentele Middle Stone Age de la fin du Pléistocène moyen en Ethiopie. Il s’appuie principalement sur une ré-analyse detrois sites majeurs de Gademotta et Kulkuletti (Région du lac Ziway, Vallée du Rift éthiopien), initialement fouillés par F. Wendorf, R. Schild et collaborateurs en 1972. Le croisement des donnés technologiques sur cesindustries à l’échelle locale permet d’aborder la question des dynamiques évolutives qui accompagnenl’émergence des premiers Homo sapiens dans la Corne de l’Afrique. Alors que les traditions techniques montrent une forte stabilité au cours du temps, notamment parl’emploi majoritaire d’une grande diversité de méthodes Levallois de débitage pour la production d’éclats et delames, se dégagent des caractéristiques techniques spécifiques qui se révèlent être des marqueurs régionaux echronologiques. Un changement majeur s’opère sur le plan conceptuel et se caractérise par une augmentation dela prédétermination des morphologies d’outils au débitage, notamment des outils phares du Middle Stone Age les outils à bords convergents (pointes). Ce changement s’accompagne d’une diminution du recours aufaçonnage et du développement de méthodes Levallois dévolues à la production d’éclats triangulaires. Encorollaire, le recours au procédé technique du coup de tranchet latéral, fréquemment appliqué aux outilstriangulaires façonnés dans les phases anciennes, diminue puis disparait au profit d’outils aux bords peutransformés. L’étude met également en exergue la maîtrise de la production laminaire Levallois, dès les phasesanciennes, antérieures à 280 ka, ainsi que la production ubiquiste de petits éclats laissés brut comme un objectifdu débitage Levallois et non Levallois à part entière. Cette étude renseigne les changements comportementaux à l’œuvre aux lendemains de l’Acheuléen, àune période charnière de l’évolution humaine, qui demeure peu documentée. Cette phase initiale du MiddleStone Age est un jalon essentiel dans la Préhistoire africaine qui annonce les phases plus récentes de la périodedurant lesquelles se multiplient les manifestations de comportements symboliques dans certaines parties ducontinent, ainsi que l’emprunt de routes de sortie d’Afrique par l’Homo sapiens. / This thesis presents new technological insights concerning Middle Stone Age (MSA) lithic industries from the end of the Middle Pleistocene in Ethiopia. Based mainly on a reanalysis of three major occupation sites from Gademotta and Kulkuletti (Lake Ziway Area, Main Ethiopian Rift Valley) initially excavated by F. Wendorf, R. Schild and collaborators in 1972, a local-scale comparison of the technology from these three nearby sites enables the evolutionary dynamics accompanying the emergence of Homo sapiens in the Horn of Africa to be discussed. While most of the technical traditions are stable through time, particularly the use of a broad diversity of Levallois methods for producing flakes and blades, several technical idiosyncrasies are also evident and can be considered strong regional and chronological markers. A major conceptual change is characterised by an increase in the predetermination of tool shape during the core reduction process, especially for the major MSA tool group – tools with convergent edges (points). This shift is accompanied by a decrease in the use of shaping techniques (façonnage) and the development of Levallois methods for the production of triangular flakes. The use of the lateral tranchet blow technique, frequently applied to triangular tools during the oldest phases, also decreases and then disappears with tools bearing little transformation becoming dominant. Furthermore, this analysis documents well-developed technical skills evident in earliest occurrences (before 280 ka) of Levallois blade production coexisting with the ubiquitous and independent Levallois and non Levallois production of small flakes. This study highlights behavioural changes evident during the still poorly documented post-Acheulean period, a turning point in human evolution. These initial stages of the Middle Stone Age are crucial for African prehistory as they set the stage for subsequent periods which see not only an increase in symbolic behaviours, but the expansion of Homo sapiens out of Africa.

Ce travail correspond à une étude technique et fonctionnelle de l’équipement lithique taillé des chasseurs cueilleurs maritimes de la région de Magellan entre 6.000-3500 et 3000 BP. Cette période se caractérise par plusieurs changements, notamment dans les industries lithiques, qui ont mené à la définition de deux unités chrono-culturelles : La «Culture d’Englefield», à la période ancienne, caractérisée par une exploitation intensive de l’obsidienne verte ; et la «Tradition culturelle Ponsonby» à la période moyenne, caractérisée par son absence et l’apparition de grandes pointes lancéolées. L’analyse technique et fonctionnelle de l’équipement lithique de trois gisements, l’un appartenant à la période ancienne, et deux à la période moyenne, a permis d’évaluer ces discontinuités techniques et de les traduire en termes de comportements humains. L’analyse fonctionnelle a nécessité d’abord l’élaboration d’un cadre analytique pour l’application de la tracéologie au contexte d’étude, car les groupes canoeros (en canot) de la région ont exploité différents types de roche: obsidiennes, rhyolites, cinérites et RGFO métamorphiques. Ces matières premières réagissent de façon différente à l’utilisation, et leur analyse pose des problèmes méthodologiques différenciés et offre des possibilités interprétatives particulières. Ceci est illustré aussi bien lors de la description des traces d’utilisation expérimentales, que de l’analyse du corpus archéologique. En particulier, l’analyse des trois gisements a donné des résultats de qualité différente et les données concernant ceux de la période moyenne sont moins éloquentes que celles du site de la période ancienne. Ceci reflète des biais dus au type de roches utilisées pour la fabrication de l’équipement et ne répondent pas nécessairement à des comportements préhistoriques. Malgré ces biais, les résultats de l’analyse fonctionnelle ont servi d’abord à mieux comprendre les activités réalisées dans chaque site (boucherie, travail de la peau, des matières végétales, osseuses et minérales…). Dans un deuxième temps, ils ont permis de contribuer à la définition de l’équipement lithique de chaque période, de discuter certains choix techniques de fabrication et des comportements de gestion de l’outillage. Finalement, une comparaison diachronique est tentée, en discutant les changements et continuités dans la composition et gestion de l’équipement lithique taillé, en rapport avec les chaines techniques de fabrication et d’utilisation. / This work corresponds to a technological and functional study of the knapped lithic tool kit of the maritime hunter gatherers from Magallanes region between 6,000-3,500/3,000 BP. This period is characterised by several discontinuities, particularly in the lithic technology, that have led to the definition of two chrono-cultural unites. First, the so called “Englefield Culture” in the old period, characterised by an intensive exploitation of green obsidian. Then, the “Ponsonby Tradition”, in the middle period, which is defined by the absence of obsidian and the onset of big lanceolate bifacial points. The technological and functional analyses of the lithic tool kit of three archaeological sites, one associated to the old period and the other two related to the middle period, contributed to the evaluation of these technical discontinuities, which were seen in terms of human behaviour. The functional analysis required, in first place, of an analytic frame for applying the Traceology to the study context. This is because the canoeist groups exploited different lithic raw materials: obsidian, rhyolites, cinerites and metamorphic FGDR (Fine grain dark rocks). These raw materials respond in a different way to the use, therefore, their analysis implies methodological requirements and special interpretative possibilities. These particularities are evidenced in the description of the experimental use wear as well as in the results of the archaeological material analysis. In relation with the last ones, the analysis of the three sites yielded different results, and the data available for the contexts of the old period is more eloquent than the data of the middle period. This difference reveals a probable bias linked to the types of rocks used for the tool kit manufacture, and does not express necessarily human behaviour. Even though the bias presented above, the results of the functional analysis helped, in first place, to achieve a better understanding of the activities developed in each site (butchering, processing of skin, vegetable, bone and mineral materials). In second place, the data improved the definition of the lithic tool kit for every period, and allowed the discussion of some technical decisions in the manufacture chaine operatoires as well as the management behaviours of the tool kit. Finally, it was attempted to develop a diachronic comparison, discussing the changes and continuities in the composition and management of the lithic tool kit and stablishing the relationships with the manufacture and use chaine operatoires. / Este trabajo corresponde a un estudio tecnológico y funcional del equipamiento lítico tallado de los cazadores recolectores marítimos de la región de Magallanes entre los 6.000-3500/3000 AP. Este periodo se caracteriza por varias discontinuidades, particularmente en la tecnología lítica, que han conducido a la definición de dos unidades crono-culturales. Primero, la llamada “Cultura Englefield” en el periodo antiguo, caracterizada por una explotación intensiva de la obsidiana verde. Luego, la “Tradición Ponsonby”, en el periodo medio, caracterizada por la ausencia de obsidiana y la aparición de grandes puntas bifaciales lanceoladas. El análisis tecnológico y funcional del equipamiento lítico de tres sitios arqueológicos, uno asociado al periodo antiguo y dos al periodo medio, permitió realizar aportes a la evaluación de estas discontinuidades técnicas, vistas en términos de comportamientos humanos. El análisis funcional requirió, en primer lugar, de un marco analítico para aplicar la traceología al contexto de estudio, ya que los grupos canoeros explotaron diferentes materias primas líticas: obsidianas, riolitas, cineritas y RGFO (Rocas de Grano Fino Oscuras) metamórficas. Estas materias primas reaccionan de forma diferente al uso, y su análisis implica requerimientos metodológicos y posibilidades interpretativas particulares. Esto se evidencia tanto en la descripción de huellas de uso experimentales como en los resultados del análisis del material arqueológico. Con respecto a estos últimos, el análisis de los tres sitios arrojó resultados de calidad diferente, y los datos disponibles para el sitio del periodo antiguo son más elocuentes que los de los sitios del periodo medio. Esto refleja probables sesgos asociados a los tipos de roca utilizados para la fabricación del equipamiento, y no necesariamente comportamientos humanos. A pesar de estos sesgos, los resultados del análisis funcional sirvieron en primer lugar para lograr una mejor comprensión de las actividades realizadas en cada sitio (carnicería, trabajo de piel, de materias vegetales, óseas y minerales…). En segundo lugar, los datos permitieron realizar una mejor definición del equipamiento lítico para cada periodo, así como discutir algunas elecciones técnicas en las cadenas operativas de fabricación y los comportamientos de gestión del instrumental lítico. Finalmente, se intentó realizar una comparación diacrónica, discutiendo los cambios y continuidades en la composición y gestión del equipamiento lítico y estableciendo relaciones con las cadenas operativas de fabricación y uso.

This study of lithic analysis shifts attention from typological studies to explicitly behavioral analyses, complimenting studies of both intrasite and intersite patterns of variability and change. Analysis of several assemblages from the Placitas Arroyo site complex reveals changing patterns of raw material procurement and selection, core reduction strategies, as well as tool production and discard. The most striking result thus far is the quite uniform emphasis on flake production from well-prepared cores, and the near absence of manufacture or maintenance of bifacial tools, especially projectile points. Associated with common ground stone artifacts, the flaked stone materials may well represent intensive food processing. Regardless, the technological patterns being revealed by this approach illustrate a productive new means to gain insights into changing behaviors in the Jornada Mogollon cultural tradition.

Le présent travail s’articule autour de plusieurs thèmes. Il s’agit de l’étude archéologique de la région du Bas Mertoutek dans la Téfedest. Cette recherche s’intéresse à tous les documents archéologiques présents sur le terrain. L’analyse des sites d’habitat néolithiques et l’étude des vestiges récoltés lors des fouilles, ont permis de les situer parmi les cultures néolithiques de l’Ahaggar et du Sahara central. De même, l’étude des monuments funéraires, l’analyse de leur distribution géographique ainsi que la fouille de certaines formes architecturales, ont conduit à affiner nos connaissances sur ces monuments et à comprendre les différents rites funéraires et les diverses positions d’inhumation. L’étude de l’art rupestre, peint et gravé, se trouvant dans ce contexte funéraire et d’habitat préhistorique, à contribué grandement à la compréhension des modalités de mise en place et d’installation de ces groupes culturels, artisans de ces industries lithiques et céramiques, des monuments funéraires et des fresques rupestres dont regorge cette région. Cette recherche a aussi contribué à mieux connaître le peuplement holocène de ces régions désertiques, ses modes de vie, son évolution et les éventuelles relations entre tous les vestiges qui ont été pris en charge dans ce travail. / The present work focuses on several themes. It is the archaeological study of the Lower Mertoutek in Tefedest. This research deals with all archaeological materials present on the ground. The analysis of Neolithic settlement sites and the study of the remains collected during the excavations allowed to situate them among the Neolithic cultures of the Ahaggar and central Sahara. Similarly, studies of monuments, the analysis of their geographical distribution and the excavation of certain architectural forms, led to refine our knowledge of these monuments and understand the various funeral rites and different burial positions. The study of rock art, painted and engraved, located in the funerary context and prehistoric settlement, has contributed greatly to the understanding of the modalities of the establishment of these cultural groups, producers of these lithic industries and pottery, funerary monuments and cave paintings abounding in the region. This research has also contributed to better understanding of the Holocene settlement of the desert regions, its lifestyle, its evolution and possible relationships between all the remains that have been taken over in this work.

Dans la Formation de Shungura l’apparition du comportement de taille de la pierre coïncide avec l’apparition de l’Oldowayen ancien et se concentre sur un laps de temps relativement resserré (Membre F et partie inférieure du Membre G ; 2,32 Ma à 2 Ma), bien que les hominidés fossiles soient présents tout au long de la séquence (3,6 à 1 Ma). Afin de questionner les interrelations et l’évolution entre les occupations humaines de l’Oldowayen ancien et les facteurs environnementaux, une approche multi-scalaire a été développée. Elle permet d’intégrer les très nombreuses données archéologiques (une centaine d’occurrences dans le Membre F et une cinquantaine dans la partie inférieure du Membre G) et paléoenvironnementales (incluant plusieurs milliers de spécimens pa-léontologiques des Membres E et F et les données géologiques de terrain). Selon le degré de préci-sion spatio-temporelle de ces différents registres de données, trois échelles d’analyse ont été rete-nues (complexe archéologique, zone d’étude, formation). L’analyse spatiale et taphonomique des données archéologiques couplée à l’analyse des cortèges fauniques a permis de démontrer que seul un petit nombre d’occurrences correspond à des occupations initiales dans le Membre F. Elles sont localisées à la base du Membre F, à proximité du paléo-fleuve Omo, dans un contexte général d’ouverture et d’aridification croissantes du milieu. À cela s’ajoute une répartition spatiale différen-tielle de certains taxons entre la partie nord et la partie sud de l’aire Type, qui pourrait être le mar-queur d’une plus grande emprise des zones humides dans le paysage de la partie sud de la Formation de Shungura. / The appearance of stone tool making behavior In the Shungura Formation coincides with the appearance of the early Oldowan. It lasts over a limited time period (Member F and the lower part of Member G: 2.32 Ma to 2 Ma), while hominid fossils are present in all members, from 3.6 Ma to 1 Ma. A multi-scalar approach was developed for assessing potential interrelations and co-evolution be-tween Early Oldowan human occupations and environmental factors. This approach has allowed to include an extensive set of archaeological and paleoenvironmental data (a hundred of occurrences in Member F and ca. fifty occurrences in the lower part of Member G, thousands of faunal remains from Members E and F, and field geological data). Depending on the spatial and temporal resolution of these different records, I have defined three scales of analysis, from archeological complex scale, to study area and formation scales. The combination of spatial and taphonomical analyses of archeo-logical data, coupled with the analysis of faunal assemblages, suggest that only few archeological occurrences correspond to primary occupations in Member F. They are all located in the lower part of Member F, in the vicinity of the Omo paleoriver, in a global context of opening up and drying of the environment. Additionally, the differential spatial distribution of several faunal taxa between the northern and southern parts of the Type area could mark a greater extension of wet areas in the southern part of the Shungura Formation.

Lithic artifacts are used in the understanding of human history. In order to do so, analyses and categorizations are made of the artifacts with the use of several methods and terminology. In this thesis, the minimum analytical nodule analysis (MANA) is used as method to analyze assemblages of lithic artifacts from 16 archaeological sites in Sweden. The sites date to Middle Mesolithic (c. 6800–5500 BC) and Late Mesolithic (c. 5500–3900 BC). The aim is to try to understand activities at the sites, and the theory of social organization is used. The results of the study indicate that it is possible to divide the sites into three types based on activity scenarios. In addition to this, the words ‘mobility’ and ‘sedentism’ are problematized in an attempt to understand how mobility and sedentism can be interpreted, both in the light of previous research and based on the results of this study. / Litiska artefakter används i förståelsen av människans historia. För att göra det, görs analyser och kategoriseringar av materialet med användning av flertalet metoder och terminologi. I den här uppsatsen, används minimum analytical nodule analysis (MANA) som metod för att analysera samlingar av litiska artefakter från 16 arkeologiska platser (eng. sites) i Sverige. Platserna är daterade till mellanmesolitikum (ca 6800–5500 f.Kr.) och senmesolitikum (ca 5500–3900 f.Kr.). Syftet är att försöka förstå aktiviteter på platserna, och teorin om social organisation används. Studiens resultat indikerar att det är möjligt att dela in platserna i tre typer baserat på aktivitetsscenarion. Därtill, problematiseras orden ’mobilitet’ och ’sedentism’ i ett försök att förstå hur mobilitet och sedentism kan tolkas, både i ljuset av tidigare forskning och baserat på resultaten av den här studien.

Na Amazônia Central achava-se que a ocupação pré-colonial resumia-se ao período dos agricultores ceramistas. Essa realidade resultava da inexistência de evidências arqueológicas de populações com economia baseada na caça, na coleta e no processamento de recursos líticos. E começou a se modificar a partir de outubro de 2001, quando o sitio Dona Stella foi localizado. Desde então, com a intensificação das escavações verificou-se que esse sítio possui uma indústria lítica diversificada, incluindo lâminas bifaciais, pontas-de-projétil e datações entre 9.460 e 4.500 AP. As prospecções mostraram que os locais ideais para encontrarmos tais evidências são as áreas de campinaranas, próximas a igarapés pertencentes à bacia do rio Negro, onde ocorram afloramentos de arenito-silicificado. Esse modelo foi testado com sucesso em Iranduba e Manaus, onde identificamos mais de vinte sítios pré-cerâmicos em areais. Ocorre que a totalidade desses sítios encontra-se parcial ou totalmente destruída e em nenhum deles encontrou-se uma indústria lítica que seja comparável tanto em densidade, quanto em variabilidade tecnológica à do sítio Dona Stella, que, por enquanto é um caso único na Amazônia Central. / Until recently it was believed that the pre-colonial occupation of the Central Amazon region could be reduced to the period of the ceramic producing horticulturalists. This reality resulted from the absence of archaeological evidence for (the more ancient) societies with an economy based on hunting, gathering and processing of lithic resources. This panorama began to change October 2001 when the Dona Stella site was found. The excavations here showed a diversified lithic industry, including bifaces, projectile points and dates between 9460 and 4500 AP. The surveys in the region since then demonstrated that the best places to find evidence of hunter-gatherers were the sand-rich areas of the campinaranas, where outcrops of silicified sandstone occur, and that can be found near the streams belonging to the rio Negro basin. This model has been tested successfully in the municipalities of Iranduba and Manaus, helping identify more than twenty pre-ceramic sites in similar contexts. On the other hand, these sites are partially or completely destroyed and none of them has a lithic industry that is comparable, both in density and technological variability, with the first one found, the Dona Stella site, which, for now, is a unique case in Central Amazon.

Os estudos de sambaquis estão entre os mais recorrentes na Arqueologia Brasileira. Entretanto, poucos se dedicaram a uma abordagem sistemática das indústrias líticas destes sítios. A presente pesquisa tem como objeto o Sítio do Mar Virado, localizado na Ilha do Mar Virado, município de Ubatuba, São Paulo. Trata-se de um sítio cemitério, onde intensa atividade de lascamento era realizada. O trabalho pode ser dividido em três partes: primeiramente buscou-se compreender os processos formativos do sítio e o histórico das ocupações regionais. Em seguida faz-se a caracterização da indústria lítica. Por fim os resultados da análise foram abordados em sua dispersão espacial, sendo a associação aos sepultamentos a principal variável testada, com o intuito de compreender se haveria diferenças entre o material lítico que acompanhava os mortos e aquele oriundo das demais atividades realizadas no local. / Studies of sambaquis are among the most recurring themes in the Brazilian Archaeology. However, only a few of them are devoted to a systematic approach to the lithic industries of these sites. The present research focus on Mar Virado Site, located in Mar Virado Island, Ubatuba, São Paulo. It is a cemetery, where intensive flaking activity was performed. The research can be divided in three parts: first it attempts to build an understanding of the site formation processes and the regional history of human occupations. Then it seeks to characterize the lithic industry. Finally, the results of the analysis have been approached in their spatial dispersion, testing the association with burials to understand if there are differences in the lithic material accompanying the dead and the lithic material from other activities undertaken by those groups in the island.

A arqueologia do Estado de São Paulo é um tema ainda pouco explorado nas pesquisas acadêmicas. Tal negligência pode ser justificada tanto pela constante transformação da paisagem quanto pelo próprio desconhecimento de seu alto potencial informativo. Essa dissertação tráz resultados inéditos para dois sítios localizados na região de Rio Claro, porção centro-leste do estado. Foram trabalhados os sítios: Abrigo do Alvo - que como o próprio nome sugere encontra-se em contexto abrigado - com duas cronologias bem estabelecidas, sendo a ocupação mais recente em torno de 1.170 cal AP e a mais antiga em 7.500 cal AP; e o Bastos - sítio a céu aberto - com a idade mais recente em 7.650 cal AP e a mais antiga em 12.640 cal AP. O objetivo da pesquisa foi o de caracterizar o conjunto artefatual dos dois sítios considerando os materiais líticos associados apenas à datação do Holoceno Médio, tendo como referencial teórico o conceito de Cadeia Operatória. Os resultados, quando comparados, mostram que esses grupos caçadores-coletores que habitaram o interior paulista no mesmo período, não compartilhavam características tecnológicas em comum, indicando grupos culturalmente distintos. / The archaeology of São Paulo State is a subject not fully explored in academic research. Such state of the art can be justified both by the constant transformation of the landscape and by the lack of knowledge of its high information potential. This dissertation brings new data from the Rio Claro region, central-eastern portion of the state. Two sites have been worked: Alvo Rockshelter with two well established occupations, the most recent around 1,170 cal AP and the oldest around 7,500 cal AP; and Bastos - am open air site - with the most recent chronology at 7,650 cal AP and the oldest at 12,640 cal AP. The aim of this research was to characterize the assemblages of the two sites considering the lithic materials dated from the middle Holocene, having as theoretical reference the concept of Châine Opératoire. The results, when compared, show that these hunter-gatherer groups that inhabited the interior of São Paulo in the same period did not share common technological characteristics, indicating culturally distinct groups.

Este trabalho apresenta os resultados da análise tecnológica de peças líticas lascadas, de dois sítios arqueológicos localizados no vale do Rio Paranapanema: o Sítio Arqueológico Vallone, no Município de Iepê-SP, Baixo Paranapanema, e o Sítio Gurucaia, no Município de Piraju-SP, Médio Paranapanema. Tal análise teve por objetivo contribuir com informações sobre indústrias líticas de sítios líticos. Acreditamos que os resultados obtidos, associados às demais pesquisas realizadas na região, possam contribuir com as reflexões do sistema de ocupação regional, haja vista que, na produção dos artefatos, cada grupo humano fez suas escolhas tecnológicas, as quais estão presentes na variabilidade do conjunto lítico e refletem as escolhas culturais, frente às necessidades que se impunham a partir da relação do homem com o meio ambiente. / This research shows the results of a technological analysis concerning chipped lithic artifacts found in two archaeological sites situated by the Valley of Paranapanema River: Vallone Archaeological Site, located in Iepê - SP, Low Paranapanema, and Gurucaia Site, placed in Piraju - SP, Medium Paranapanema. Such analysis aimed at providing some information about lithic industries of lithic sites. It is believed that the results of this research, joint with others carried out in the same region, can contribute to the reflexions of the regional occupation system, since each human crew, when producing the artifacts, made their technological choices, which are existent in the variability of the lithic ensemble and reflect the cultural choices made based on the needs that come up in the relationship between the man and the environment.