Les objectifs de ce travail sont doubles. Le premier est la caractérisation de l’organisation technique, économique et sociale de la métallurgie du fer au sein d’un vaste ensemble géographique. Le second est de vérifier l’existence de permanences ou de ruptures chronologiques et géographiques dans ce système. Une chronologie large (450 a.C. – 100 p.C.) incluant la conquête romaine des territoires concernés permet de vérifier si les changements politiques influent sur les productions comme celle du fer. Dans un premier temps, l’étude des structures de production, ainsi que des outils et des déchets de travail du fer, fournit des éléments permettant de restituer les procédés techniques employés par les métallurgistes. Il a ainsi été possible de démontrer que certains forgerons étaient spécialisés dans la fabrication d’un panel réduit d’objets, tandis que d’autres avaient une activité beaucoup plus large. Dans un second temps, un travail sur l’organisation des espaces de production a permis de démontrer l’existence de plusieurs types d’arrangements des structures au sein des ateliers. Ces éléments, ajoutés à ceux sur les techniques de fabrication, conduisent à déterminer l’existence de lignées techniques : des personnes utilisant des structures similaires pour effectuer des productions similaires. Enfin, la réinsertion de ces données dans leur contexte social, en particulier en associant chaque production au statut du site sur laquelle elle est effectuée, démontre de fortes variations régionales et chronologiques de l’organisation du processus métallurgique sur la façade atlantique. Au début du second âge du Fer, la métallurgie est peu répandue. Elle ne produit que peu de déchets et s’effectue en contexte rural. Les quantités de métal les plus importantes sont travaillées au sein des domaines des élites. La fin de la Protohistoire est marquée par une augmentation de la quantité de fer produit et par la spécialisation toujours plus marquée des activités, démontrant une forte séparation des diverses phases de travail. De plus, la proportion de travaux de forge au sein des agglomérations augmente. Autour du changement d’ère, la quantité de fer produit croît encore, de même que le nombre d’espaces de travail spécialisés. Cela atteste de la prolongation des tendances amorcées des périodes antérieures. / The objectives of this work are twofold. First, to characterize the technical, economic and social organization of iron metallurgy within a vast geographical area. Secondly, to verify the existence of permanence or chronological and geographical breaks in this system. A broad chronology (450 BC - 100 p.c.) including the Roman conquest of the territories concerned, makes it possible to verify the influence of political changes in iron production. First, the study of production structures, as well as iron working tools and waste, provides elements for restoring the technical processes used by metallurgists. It was thus possible to demonstrate that some blacksmiths were specialized in the manufacture of a reduced panel of objects, while others had a much larger activity. In a second step, work on the organisation of production spaces made it possible to demonstrate the existence of several types of structural arrangements within the workshops. These elements, added to those on manufacturing techniques, lead to the determination of the existence of technical lines: people using similar structures to carry out similar productions. Finally, the reintegration of these data into their social context, in particular by associating each production with the status of the site on which it is carried out, shows strong regional and chronological variations in the organisation of the metallurgical process on the Atlantic coast. At the beginning of the second Iron Age, metallurgy was not very widespread. It produces little waste and is carried out in a rural context. The largest quantities of metal are processed in the elite domains. The end of Protohistory is marked by an increase in the quantity of iron produced and by the ever more marked specialisation of activities, demonstrating a strong separation of the various phases of work. In addition, the amount of smithing works increase in urban areas. Around the change of era, the quantity of iron produced increased further, as did the number of specialized workspaces. This reflects the continuing trends initiated in previous periods.

Cette thèse est le résultat de trois années d’études sur le site archéologique de Sangtarashan situé à l’ouest de l’Iran au cœur de la chaîne montagneuse du Zagros, dans la province du Luristan. Le site, découvert en 2002 et fouillé pendant six campagnes de 2005 à 2011 a révélé une situation très particulière aux archéologues. Au sein d’une structure architecturale circulaire, en pierre, chevauchée par plusieurs autres structures, plus de deux milles objets et notamment plusieurs centaines d’objets métalliques connus sous le nom de Bronzes du Luristan ont été mis au jour. Certains étaient en lot et d’autres étaient isolés parmi les blocs de pierres des constructions ou éparpillés sur toute la surface du site. L’enjeu de cette étude était en premier lieu de proposer une fonction et une datation pour ce site et de là, le contextualiser dans son milieu géographique et historique. L’étude architecturale et l’étude spatiale des objets nous ont conduit à suggérer que Sangtarashan ait été un lieu cultuel où le dépôt d’objets métalliques constituait une coutume réalisée par des fidèles, potentiellement des voyageurs ou des nomades. L’étude typo-chronologique de plus d’un millier d’objets, en grande partie métalliques, nous a permis de situer l’occupation du site à l’âge du Fer avec deux phases consécutives différentes : l’âge du Fer I-II pour la première phase et l’âge du Fer II-III (même IV ?) pour la seconde phase. Il semble que la coutume du don de la première phase prenant la forme de dépôts en lots d’armes et de vases se soit transformée, à la seconde phase, en un don d’objets isolés, de taille plus petite et de nature plus variée. L’hypothèse d’une fonction non cultuelle (domestique) pendant la seconde phase n’est pas totalement écartée au regard de la prolongation des structures architecturales vers l’ouest. De futures fouilles archéologiques pourront confirmer ou infirmer cette hypothèse. La richesse du mobilier de Sangtarashan fait de ce site une base de référence pour les prochaines études des Bronzes du Luristan. Elle nous a permis également de proposer des datations pour certains types d’objets jusqu’à présent uniquement attestés parmi les objets de collections. Le fin mot de l’histoire, Sangtarashan semble être, en parallèle à Sorkhdom-é Lori, le deuxième sanctuaire de l’âge du Fer de la région du Zagros central où les fidèles avaient une prédilection pour le don d’objets métalliques. Même si cette étude a permis d’éclaircir plusieurs points concernant la région du Luristan à l’âge du Fer et ses Bronzes énigmatiques, de nouvelles questions ont parallèlement été soulevées méritant d’être étudiées par de nouvelles recherches. / This dissertation is focused on the archaeological site of Sangtarashan located on the western Iran, in Luristan. Discovered in 2002 and excavated during six campaigns from 2005 to 2011, the site had revealed, more than two thousand objects particularly several hundreds of "Bronzes of Luristan", in a circular architectural structure, in stone, overlapped by several other structures. Some of them were deposit as packages and others were isolated between stone blocks of walls or scattered over the entire surface of the site. The aim of this study was to propose a function and a dating for the site and then, to contextualize it in its geographical and historical environment. The architectural study and the analysis of spatial organization of objects led us to suggest that Sangtarashan was a ritual place where the act of deposition of metal objects was a custom by prayers, potentially travelers or nomads. The typo-chronological study of the objects allowed us to date the occupation of the site to the Iron Age with two consecutive phases: the Iron age 1-Il for the first phase and the Iron age Il-III (even IV?) for the second phase. lt seems that the form of dedication change from the first phase to the second one. In the first phase objects are dedicated as deposits in packages of arms and vessels. While in the second phase, they take the form of deposit of isolated objects of smaller size and of a more varied nature. The richness of the Sangtarashan's finds makes this site as a reference base for the next studies on the Bronzes of Luristan. It also allowed us to propose some dating for several types of objects till todays only attested among the objects provide from the illegal diggings.

Moving away from the highly regionalised and constrained purely humanistic and empirical studies of hillfort location and morphology, this study is a multi-regional GIS-based analysis of the form and siting of several groups of hillforts across Britain. Hillforts in Dartmoor, Aberdeenshire, The Gower and Warminster are assessed, four regions that are topographically diverse. The highly varied topography of these regions also tests the GIS-basis of this study, another important intrinsic aspect of this novel research. GIS-based analysis has never before been applied to a study of hillfort location and morphology to this degree and, as with any innovative methodology its worth has to be tested and assessed. The thesis demonstrates that GIS-based analysis, when combined with field visits, provides a fundamental insight into the possible influences of hillfort location and morphology, which fieldwork alone will never be able to do. The GIS-based analysis developed here focuses largely upon examining degrees of movement and visibility. Unlike other GIS-based analyses of movement and visibility this integrates the two to examine visual pathways across landscapes to further investigate the visual qualities of hillforts within the various test areas. The study demonstrates that GIS-based analysis when combined with fieldwork can be affectively applied to qualitative based questions surrounding hillfort location and morphology. The overall results of this analysis had some relatively predictable results whilst there were some very surprising cases. A large number of entrances were placed within the most accessible area, however in the case of Battlesbury there was evidence for the complete disregard to accessibility within the orientation of its northwestern entrance. There were also numerous examples of the placement of a site's most prominent morphological components in correlation with the blind pathways. In these cases sites were orientated to encourage an element of surprise upon the approaching travellers.

Sandby borg, an Iron Age ringfort on Öland, Sweden has been and is still at the center of attention in media and archaeological research. The massacre uncovered at the site during recent years opens many doors for analyses on the Migration Period (c. 400-550 AD) Iron Age skeletal remains. Eighteen teeth (molars) from 12 individuals and three rodent teeth were chosen for strontium (87Sr/86Sr) analyses. This was done to establish whether these individuals were locals or non-locals to Öland. The analyses displayed a, for the most part, local 87Sr/86Sr ratio. Two non-locals were identified, as well as a pattern of higher 87Sr/86Sr peaks on numerous of the individuals during a certain age span, which could indicate a local weaning process with a special food. / Sandby borg

This thesis primarily discusses the development of late Iron Age Saami settlement patterns in greater Norrland's forest area, from the establishment of the Settlements through to historical times. The Settlements are chiefly characterised by hearths, but it is also important to study Saami sacrificial sites when trying to understand the significance of the settlement patterns. Central to the thesis is how the archaeological material can be applied to questions concerning the introduction of reindeer herding. During the early Viking period a significant change in the settlement pattern of greater Norrland's inland occurs. New niches start to be exploited, moving away from the earlier shore-bound model. The Settlements are relocated to areas with good reindeer grazing land, by small streams, bogs and small lakes. The principal features are concentrations of hearths, which arise in large numbers, most often in groups of three to ten. A number of the artefacts found at the settlement sites are also found at Saami sacrificial sites from between 800 to 1350 AD, suggesting that the hearths should be studied in the context of Saami culture. A wide range of artefacts have been discovered during excavation of the Settlements, which suggests extensive contacts, mainly to the east and the Ladoga area, but also with Norway to the west. The artefacts display a continuity from the Viking period into the 1700's, and the dating of the hearths show a similar chronological spread. The study area has supported a reindeer herding forest Saami society during historical times, the settlement pattern of which has close similarities to that found under the Viking period. This implies that the settlement pattern that emerged during the Viking period can probably be related to an emergent reindeer herding system. Reindeer herding was undertaken in combination with hunting and fishing - so called semi-nomadism. It is suggested that the forest land Saami society become so dependent on reindeer herding during the Viking period, that it controlled the settlement pattern. / digitalisering@umu

This thesis examines the wetland sacrifices that were performed in Northern Europe in the Iron Age. Skedemosse on central Öland is the largest wetland sacrifice in Sweden and was the site of a cult which sacrificed animals and humans. Between the late second century and well into the fifth century the place was also used for large sacrifices of military equipment. New radiocarbon dates has shown that the place functioned as a ritual place from the Pre Roman Iron Age and into the Late Viking Age. Both in the Iron Age and later wetlands seem to have been both venerated and feared and the thesis discusses why this came to be, and how it can be seen in the archaeological material. A smaller part of the sacrificial site of Skedemosse was selected for a closer study and it was possible to establish several depositions which appear to have been treated slightly different from each other. The investigations of the animal sacrifices have focused on the horses as they are the most common animal. The horse was an important mythological animal in the Iron Age and they were equally important in the cult. The horses in Skedemosse were eaten in ritual meals, and it is possible that some of them took part in ritual races along the ridge east of the former lake.  Such races were called skeið and the name Skedemosse may be derived from this word. Skedemosse is also rare because the remains of ca 38 people have been found in it. Some of these people have suffered a violent death. They are compared to other bog bodies from northern Europe and the follow a similar pattern to those; In the Pre Roman Iron Age mainly women and children were sacrificed and after the first century AD mainly men ended up in the lake.

Jag har gjort en jämförande studie mellan utvecklingen i Danmark och utvecklingen på Gotland under romersk järnålder. Som underlag har jag framförallt använt Lotte Hedeagers avhandling Danmarks jernalder – Mellem stamme og stat (Hedeager 1992) och Kerstin Cassels redovisning av det arkeologiska källmaterialet från Gotland i hennes avhandling Från grav till gård – romersk järnålder på Gotland (Cassel 1998). Syftet med studien var att undersöka om Hedeagers slutsats att det etablerades en centralmakt med territoriell kontroll, en tidig stat, i Danmark redan i övergången mellan äldre och yngre romersk järnålder, även skulle gå att dra vad gäller Gotland baserat på det gotländska källmaterialet, och i så fall om detta kunde förklara uppkomsten av stengrundsbebyggelse och fornborgar. Jag har jämfört det arkeologiska källmaterialet från Danmark och Gotland avseende gravfynd och offerfynd, bebyggelseutveckling och landskapsutnyttjande samt vapenoffer och försvarsanläggningar. Vad gäller gravfynden går det att se stora likheter både i hur gravgodset ser ut och i de förändringar som sker mellan äldre och yngre romersk järnålder. Det går också att se stora likheter i kulturlandskapets utveckling med en övergång från ett extensivt åkerbruk med kollektivt brukande till familjejordbruk med intensiv odling. Både på Gotland och i Danmark kan man datera storskaliga försvarsanläggningar till slutet av äldre eller yngre romersk järnålder. Exempel på detta är Olmerdiket och Torsburgen. En väsentlig skillnad mellan Danmark och Gotland är hur praktföremål fördelas mellan offerfynd och gravar övertiden från förromersk järnålder till folkvandringstid. Det jag kommit fram till är att likheterna är så stora att man genom att tillämpa Hedeagers analysmodell kan dra slutsatsen att omläggning av jordbruket, uppkomst av stengrundshus och stensträngar, och anläggande av försvarsborgar gjorts på initiativ av en centralmakt med kontroll över det gotländska territoriet.

In this study bone material from six stone ship settings and a total of seven deposits of bones from two sites on the island of Gotland have been analyzed. Four ship settings from the burial site at Gålrum in Alskog parish and two from Tängelgårda in Lärbro parish. The ships contained both cremated remains and inhumations.  Human remains were identified in five of the analyzed ship settings and a minimum of six individuals was identified in total.  Animals were found in three of the ship settings, where one of them contained only the burned remains of a dog. Two of the ships contained inhumations, one in Gålrum and one in Tängelgårda. The one from Gålrum was determined to 17-19 years of age but of undetermined sex. The inhumation from Tängelgårda was determined to a male of 35-64 years of age.  Of the cremated remains none could be determined to sex but all were determined as adult individuals. The temperature that the cremated bones had been exposed to during the cremation was similar between the different ship settings, with one exception which had been exposed to a slightly higher temperature. The results of the material have also been compared with those of other osteological analysis from ship settings on Gotland. The interpretation of this has been that the amount of bone and the number of individuals that has been buried in stone ship settings vary from different sites. Some ship settings have contained several burials and have been interpreted as family graves (Pettersson 1982) which do not fit with the results from this analysis where all the ship settings contained a maximum of one or two human individuals, so the results from the ship settings on Gålrum and Tängelgårda differ from earlier interpretations of ship settings on Gotland.

Is it relevant to treat bog bodies as its own category within the archaeological research? This paper examines how scholars and researchers, through their publications and depictions of the preserved prehistoric individuals, are helping to create such a category. What has changed in the way scholars look at and describe these astounding finds? Also examines the various theories researchers provide regarding bog bodies and how they are deposited.

Thesis (doctoral)--Uppsala universitet, 2002. / Includes bibliographical references (p. 399-435).