dAf 40
COLARDELLE Michel, VERDEL Éric
Les habitats du lac de Paladru (Isère) dans leur environnement
La formation d'un terroir au XIe siècle
This study, by M. Colardelle, E. Verdel and a team of forty researchers, of a fortified settlement constructed in 1003 AD and abandonned before being submerged under the water of lake Paladru around 1040 AD, enlarged to cover the surrounding landscape, has made an essential contribution to archaeology and medieval history. Besides fundamental information concerning. architecture in wood, arts and crafts, food supply and agriculture, of which the evidence was remarkably well preserved, this research has led to a new interpretation of the events leading up to the revolution of 1000 AD, a key period in the history of the West. Exceptional, considering the investigational techniques used (stratigraphical excavation in subaquatic conditions, environmental sciences and archaeometry) and the results obtained, this study is of interest to historians and naturalists, but also to those interested by the relationship between archaeology and history, the first producing unique evidence to support the second.
Abstract
Abstract
1. Introduction
Paladru lake, of glacial origin, is near the great Alpine upland of Chartreuse, rising to 500 metres amid the morainic mountain in the eastern region of Terres Froides. Heavy and stoney, the soil is largely infertile, and with a continental septentrional climate marked by heavy rainfall (an annual average of 1160 mm) and high winds, the region is relatively inhospitable for much of the year.
The area mostly covered by forests, coppices and pasture land, resources traditionally devoted to an agropastoral economy dominated by cattle-raising. Nevertheless, since the Middle Ages, Fure valley (an emissary of the lake), has seen the development of an active and high quality metallurgical industry, which has benefited from nearby hydraulic power resources, an abundance of wood, and the proximity of Chartreuse's iron ore.
Historically Paladru lake, for a long time, remained apart from the great settling movements. There were temporary settlements in the late Neolithic .era, circa 2750 BC (notably Baigneur's village), but in general its banks were only sporadically settled during the Roman period, probably due to the inhospitable nature of the soil and its distance from the main communication routes. Such was the case even during the late Middle Ages when the region belonged, from the 9th century on, to the earldom or pagus of Sermorens who established a centre of power in Voiron, 15 kilometres away in the same place as a vast gallo-roman villa of the Early Roman period.
This territorial entity dismembered in 1107, when, its ownership having been hotly disputed by the archbishops of Vienne and the bishops of Grenoble since the end of the 11th century, it was partitioned by Papal arbitration. Of the twenty-two castellanies that made up the territory, Paladru, Virieu, and Clermont were among those given to the archbishopric of Vienne. This text, the first to make explicit reference to Paladru, formally established the prior existence of the lakeside seigneuries. Their construction around the mottes, which appeared in the years 1040-1050, hence follows the abandon of the lakeside fortified settlements, between 1030 and 1040. The paucity of written sources about this period, however, makes it impossible to elaborate on the circumstances surrounding the emergence of the first castellan lineages, which archaeology alone can properly document.
The location of a few lacustrian stations has long been known ; regional historians of the 16th and 17th centuries sometimes mentioned them. Local popular traditions also speak of ancient "cities" submerged in the past. It was only in the second half of the 19th century, however, that the first scientific investigations. began. These were soon followed by surveys on the site of Grands Roseaux at Paladru. These surveys uncovered a number of wooden, ceramic, and metal artifacts dating from the "Carlovingian" era. However, the first researchers still believe that the stakes driven into chalk served to support elevated platforms and interpret the groups of piles as vestiges of lake dwellings.
The two sites at Charavines, Baigneurs and Colletière, were not identified until the beginning of the 20th century, by Hipppolyte Muller. He made some observations about the medieval settlements, and believed them, quite logically, to be contemporaneous with Grands Roseaux.
ln 1971, a lakeside development project threatened both the Neolithic and Medieval sites, so rescue excavations were launched. Colletière field received limited funds unti11986, when it began ta benefit from a multi-year authorization from the Ministery of Culture and from the signing of an agreement by the State, the Isère Department and the Town of Grenoble which provided funding ta permit a systematic analysis. Similarly, since 1985, "the Programme pluriannuel en sciences humaines Rhônes-Alpes" has financed the paleo-environmental research.
Underwater sites offer the advantage, thanks to the remarkable conservation of the organic remnants ordinarily destroyed on terrestrial sites, of concealing exceptionally rich archaeological resources. Moreover, Colletière offers the advantage of not having been occupied after the year 1000 and of not having had any significant modification of its deposits. It merited, therefore, whatever appropriated excavation techniques could be put in place. After several years devoted to establishing the excavation methods appropriate to the site, in cooperation with the prehistorians who were exploring the neighbouring village of Baigneurs, the proper procedures have been adopted.
When the bathymetry is sufficiently high, divers install in the selected area a triangular metal frame, 5 metres long on each side, and positioned according to the general layout of the site. Each large triangle is subdivided into 25 smaller metrical triangles which constitute the excavation unit proper for scaving and sampling, and from which all counting’s and distribution maps are made.
Sedimentary columns must be previously extracted from several points to reveal the stratification profile and to permit sedimentologic laboratory analyses. The removal of archaeological layers is done manually and all sediment is saved in buckets for sifting to improve visibility, which is often poor, a system of electric pumps creates an artificial stream of water ta wash away suspended particles.
During the period of lowest water-levels, dry scavings are sometimes possible on the highest part of the site. The methods used are the same but we can, simultaneously, cover the largest surfaces while paleoenvironmental specialists themselves determine the complementary samplings they think useful.
Field studies and then resulting interpretations attain in this case an incomparable quality.
Treatment of archaeological sediment includes several phases : sifting in water with various meshes retrieving whole or fragmental artifacts not seen on the site and petrographic selection and weighing of pebbles, fragments of cooked clay and charcoal, and an exhaustive collection of the paleoseeds or vegetative macroremnants, and bones of terrestrial or aquatic fauna. It is also at this stage that the artifacts are inventoried, numbered and then conditioned before being sent to either study or treatment laboratories.
2. Soil and climate
Among the different disciplines which have contributed to the study of the lacustrian area, sedimentology and diatoma determination have both shown that water-levers of Paladru lake have varied greatly over the last 2 000 years.
Lower by two or three metres in the 2nd and 3rd centuries, it climbed close to its present height before it fell again in the 10th century. This lowering, by drying the littoral shallowers, was the beginning of the establishment of settlement in the year 1 000. The meteoro-climatic origin of these bathymetric oscillations seems hereafter clearly established, at least for the 10th and 11th centuries.
Visible in the quasi-totality of sedimentary columns, a characteristic stratum distinguishes itself by its very elevated level of carbonates which resulted in the acceleration of the process of precipitation of dissolved chalk, encouraged by the reheating of the surface waters. On the other hand, the 16O / 18O ratio, measured in the growth rings observed on an architectural stake at Colletière, indicates an evapotranspiration more important today, especially when interpreted in terms of an hydrous deficit provoked by a diminution of average precipitation in the second half of the 10th century and the beginning of the 11th. It is also equally probable that the simultaneous abandonment of lakeside stations, thirty years after their establishment, is due to a new transgression which progressively covered them again. It is certifiable, in effect, that the anthropic sediments at Colletière and Grands Roseaux are, without transition, covered by lacustrian chalk.
Paralleling these alluvial variations, sedimentology has also established the presence, between phases of mineral deposition of multiple organic layers, of which some have been dated to the High Middle Ages by radiocarbon. Richer in phosphates, they show evidence of the first attempts at cultivating the soil at the end of the Merovingian period or in the course of the Carolingian period. These sporadic clearings of the ground would have been, nevertheless, of a limited scope and duration. However, the clearings of the ground of the Year One Thousand is a completely different story. They would radically and definitively influence the lakeside area, which would then see the non-stop exploitation of the catchment basin until the present day.
Paleoecological studies have therefore enlarged their field of research to terrestrial investigations to try to restitute, as faithfully as possible, the landscapes of the 11th century both before and during colonization. To this end, a pedological inquiry has evaluated global agricultural potential and determined the principal categories of soils as a function of their geological origin, geochemical composition and ground texture. The evolution of vegetation has been examined by comparing present and former characteristics, known through phytosociology, determining on pollens, paleoseeds and vegetative . macroremnants preserved in lakeside sediments since before the year 1 000 and trapped in archaeological layers. Examination of the results of these various testing procedures confirms that the soil of Paladru was, before the construction of medieval houses, largely occupied by a massive primary forest. The clearings undertaken over several decades were important, the deforested zones carefully selected to respond to the need for construction wood, and to create parcels of land that would promote a diversified cultivation of cereals. The methodical use of gasses naturally present or introduced by human efforts (gathering and arboriculture) indicates that colonization followed rational ways, adapting remarkably well to agricultural processes and to agroalimentary needs.
The same observation holds equally for the raising of domestic livestock, above all porcine, which uses at the best forest resources. The good sanitary conditions of the herds and the clearing systematically undertaken in view of achieving the best return shows again the good adaptation of pastoral techniques. If hunting game was resorted to only occasionally, fishing, on the other hand, judging by the numerous fish remains and the abundance of equipment, was regularly practiced, above all between seasons (spring and autumn) to complete the diet.
3. Colletière settlement at Charavines
Today covered again by water, the Colletière settlement, constructed not on pilings but directly on the ground, was built on a lacustrian chalk beach, forming a peninsula. The only apparent vestiges are a pleiad of several hundred oaken posts and a few plank outlines. At first sight chaotic, the distribution of the bearing elements in fact obey a fairly strict measure. Indeed, diameters, orientations and intervals correspond to regular cyclical measurements.
Inside an approximately rectangular shaped defensive fence, which encloses 1300 m2 of surface, the large inhabiting buildings were erected and two of them were excavated. Around them were stabilizing devices made out of horizontal beams intersected and superposed, driven into the lacustrian chalk. These houses were erected according to a pre-established plan and had clear differenciated functions.
in the center was House 1, which served as a place of residence for the leading family, and hence was larger and higher than the others. House 2 , to the South, sheltered people of a more modest social condition. Open sheds and outbuildings, used for the domestic livestock stablizing, were built on the inhabiting buildings. Between houses and the defensive fence were also artisanal workshops placed in open spaces and light constructions.
Sedimentological analysis were done systematicaly into the archaeological layers and confirmed that they were occupation strata, produced by accumulation of both vegetal scatterings and diverse refuses.
In order to specify the excavation observations, many physico-chemical parameters were drawn, like for example, the phosphate amount, the ratio between inorganic and organic proportion and the deposit granulometry. These various data, when added to the classical stratigraphic analysis, structure recordings and artifact distribution, allowed us to identified the settlement's spatial functions, for its thirty years of occupation.
Extended to the study of the main classes of artifacts, the same multidisciplinary process leads us largely beyond a simple typological classification. In this, metallographic analysis shows a remarkable technological knowledge (soft iron and steel uses, edge hardening, etc.) when applied to the very numerous forged tools and other extractive or transformative metallurgical by-products. Dendrological determination and track analysis of wooden furniture’s, most frequently used raw material, also revealed an excellent knowledge of tree species whose specific qualities were selectively exploited in regard with the make up of the artifacts.
The social differences between the Settlement inhabitants could be confirmed by the identification of musical instruments and game pieces. Some inhabitants play chess and do some kind of a "skillful" music with "elaborated" musical instruments, while others play backgammon games and what we would call "popular" music. The abundance of leather proves that hamessmaking and shoemaking were done with the livestock skins, probably directly on the site. Among leather snippets, which are still to be listed and described, we find mainly from the settlement dumps : breeches, hamesses, belts and crossbelts.
If we only refers to artifact's typology, Colletière's settlement would be dated, broadly between 9th to 12th centuries. Silver coining, whose relative quality rightly shows a renewal of the economic currency, also shows a more restricted period of time, since the older coins date back to the end of the 10th century and the more recent ones to the years 1030. However, we know how variable a numismatic datation can be if we do not take notice of the coins circulation duration.
Radiocarbon, whose uncertainty margin has been reduced by the number of measurments, gives us matching informations by dating the occupation between 994 and 1021. Dendrochronology was still uncertain a few years ago because we lacked a published local scale that could be checked. We now know that the trees used for construction were cut in 1003-1004 and that the last-repairs were completed in 1034. This terminus ante quem... which we will be carefull not to consider as final before the archaeological excavation ends, can be retained since it relates to our results about the lacustrian level variations during the same period. Indeed, indicators that are difficult to refuse, are suggesting that meteorological changes were undoubtedly at the origin of the abandonment of Colletière, which was progressively submerged by water and deserted c. 1034-1035.
4. The other sites
At the Roman period Paladru's region was part of Vienne's city territory which was part of the Narbonnaise province, conquered in 121 BC. But it is only from the beginnings of the Roman Empire that the first remnants of occupation on the lake's banks are dated. As through it was only either modest peasant settlements inhabiting the bottom of small tributary talwegs or other fisheries on the lake's banks. None of these settlements seem to be posterior to the 3rd century and at this time we know of no site from the Merovingian period. Some exception are the villae, founded at the western limits of the basin's slopes, in Bourbre's valley, whose soils and exposure are more propitious to agriculture.
Except for a few signs of a temporary and modest exploitation during the 9th century, which has been identified by analysis of lacustrian sediments and pollens, it is in fact at the beginning of the 11th century that the region is definitely settled and developped. Apart Colletière, two other contemporaneous lakeside stations are known. Pré d'Ars in Le Pin and Grands Roseaux, both at Paladru lake. Trial trenches were made at the end of 19th century by E. Chantre on Grands Roseaux's site and have delivered many artifacts (in wood, metal and ceramic). Typologically identical to those from the Colletière site, they confinn, as architectural remnants, a similar cultural phase for its inhabitants. The Pré d'Ars station engulfed since 1870 at 20 metres depth, was also of a similar type. However, since this site was partially constructed on a higher bank, and thus protected from lacustrian transgressions, Pré d'Ars was inhabited until the 13th century.
Accidentally discovered at the bottom of the lake, a monoxyleous pirogue, remarkably conserved, dates from the 14th century. Its technical study was important since this kind of Medieval boat is hardly known. Moreover, common use of similar pirogues has been proven indirectly at Colletière around the Year One Thousand by the presence of both inland water transport material and fishing nets.
Paladru's region and its surroundings contain a fairly dense network of fortified mottes, erected on a collineous relief. The most voluminous and better preserved, "Châtelard" fortified motte in Chirens, was the subject of an extensive excavation, while others had trial trenches or surface surveys. Artifacts from these mottes, although considerably Jess abundant, are very similar to those from the lakeside stations. However an elaborate study of shapes and pottery designs collected at "Châtelard" suggests a slight chronological shift from the other lakeside site mentioned above. The current hypothesis is that the building of the fortifications (none of which is documented by archives) would be slightly subsequent to that of the three lacustrian sites, maybe even contemporaneous to their abandonment (1030-1040). In this case, their building would correspond to the first lines of castellans which were emerging in the same time and replacing the reigning powers of the domanial centers established on the lake's banks. The occupation phase of some mottes is short and does not extend beyond the years 1070. These early withdrawals certainly reflect a quick concentration of powers and the reduction of castellanies in number, as they were swallowed by more powerful neighbours. But other mottes still continue to be occupied by families who managed to remain in place. Since the beginning of the 12th century, these castra are mentionned by many texts. Castra which do not specify if they are either earthen fortifications or stone castles which remain today, like Château- Vieux in Paladru or Clermont in Chirens, do not show any elements of architectural masonry from before the 13th century. Should we conclude that some types of mottes lasted until that date, when they were reconstructed in stone? In fact, only systematic excavations would reveal whether or not there exist transitional states, traces of which no longer remain in elevation.
If, in the near Bourbre valley, the first private churches and chapels go back to the First Middle Ages (7th-9th centuries), constitution of the parochial churches network around Paladru lake certainly followed the incastellamento movement which characterizes the 11th century. This emerging network was already contribuing to framing and sedentarizing settlers on the newly constituted territories. Unfortunately in this region, the Religious Wars, the Counter-Reform and the French Revolution were each in turn causes of destruction, partially or entirely, of the medieval religious buildings. Many of them were completely reconstructed in the 17th century, because the pre-existing buildings, either too damaged or too exiguous, were razed. This is why, with few exceptions, we have to confine our research to archival sources. Although rare before the 13th century, they allow us to know the construction dates and the approximative location of most of the churches. However, it is another story for monastic foundations, either Benedictine or Carthusian. If the Carthusian monastery of Silve Bénite in Le Pin, founded in 1116, was destroyed at the end of the 16th century, it was de novo erected not far away from its primary location, leaving then its main remnants for further excavation. The case of the Cluniac priory Notre-Dame-du-Gayet in Chirens, probably erected through the generosity of the first lords of Clermont, goes even further in this way indeed, even though it underwent modifications at the end of the Gothic period, it still illustrates a good example of Romanesque architecture from the Dauphiné of the second half of the 11th century.
5. Interpretation and research directions
Outside of the major settlement zone during Antiquity, Paladru lake presents mainly discreet and transitory traces of occupation, inspite of a slight climatic warming - signs of which are also found in the vicinity and in Europe. This reheating lowers the lake level and allows the installation of some small establishments on its banks. But this weak and diffused settlement does not seem to increase during the High Middle Ages.
During the 10th century, after the cooling of the Low Empire and the raising of its water level, the climate warms up again and, by long pulses of 30 years, provokes a new lowering of the lake. Here again, observed phenomena from many disciplines at the Paladru lake (sedimentology, isotopic determination, etc.) strengthen and enlighten, by their chronological and quantitative accuracy, other hypothesis which were formulated from scientific studies but also from the rare existing written sources. Observation of this phenomenon, which takes place before the sufficiently important colonial settlement on Paladru lake's banks, but also of the group of the pre-Alpine average mountains and many European "deserts" in the beginnings of the 11th century (here 1003-1004), thus authorizes us to take a position in the debate about the origins of the "Year One Thousand Revolution". Indeed, there does exist a favourable climatic phase, in the 10th century, which could explain by the improvement of agricultural production, and therefore of the alimentation and economic profitability, the demographic growth and as well the social evolution which characterize this period of time.
The littoral settlement of Colletière gives sure evidence about this mutation. Part of a clearing of the ground movement sufficiently important and consistent to suggest - without stating it positively, because of missing written sources - that it was a planned expansion, this settlement seems to be hybridal - or rather a mutant. Constructed in wood, cob and thatch, it takes the shape, besides being technically excellent, of a very organized and symetrical unit, presenting, inside a large rectangular fence located in a peninsula, a main single-storeyed building with chimney planked by horse stables and two other buildings. The middle building, as we can judge by the domestic artifacts that are known through its inhabitant remnants, sheltered the leading family (masters of an agricultural domain whose fortified curtis or villa is the center ?). We can interpret it as an aula. However, we have to add that the social status of the inhabitants from the two other buildings distinguishes itself only in details. Indeed, all of them are farmers (cereals and fruits basically) and stock breeders (pigs, goats and sheeps, Bovidae) and they all woodwork, weave, leather-work, horse ride and bear weapons. Inside the internal spaces, the littoral area is divided into two parts. One is devoted to forges and probably to iron furnaces (fabrication of knives, weapons, horse accessories, tools), while the other is for maintenance of pirogues, fishing nets and horse shoeing. Cattle was also parked for the night inside the fence, to the hill side.
The forest, essentially of the mixted oak grove type, which spreads on the neighbouring hills, had been quickly cut out to leave place for agricultural lands. By using to the best their natural qualities which are very contrasted (types of soils, sun and wind exposure, etc.), we find cereal distribution like rye, wheat, oat, barley and millet, but also orchards, grasslands and small gardens (essentially for hemp, a few peas, beans and lentils). If nothing can allow us to prove the rythm of rotation cropping, many signs indicate a two-field system. If Bovidae and horses were put to pasture in grasslands, pigs on the other hand, were having the benefit of the acorn "crop" in the forest while Caprinal could have the edges and sides of roads. Alimentation was completed with fishing, mainly practiced during spring and fall, and by the gathering of a large variety of wild species. Hunting, of small aquatic game essentially, only afforded an extremely modest added contribution.
Artifacts collected in houses and the dump is unusually abundant, rich and diversified. This relates certainly to its inhabitants social status (peasant knights, like contemporaneous texts from other regions allowed us to imagine them), but also to the conservation conditions (water immediately covered again the settlement, sealing it definitively, hence assuring the preservation of organic matter). These artifacts give anyway an excellent vision of the technical level of that period, of its quality of life and also of the relative portion of self-sufficiency. Although Colletière settlement was largely autarkical, it had nevertheless specializations - agriculture and breeding, metallurgy, tawing - and also traded in order to obtain certain types of artifacts (turned wood, potteries). Coins, silver deniers and oboles, were effectively rather abundant, sure signs of wealth if we judge by the contrast given us by written sources about the Rhône peasants of this time.
Everything hence indicates that, with two other contemporaneous lakeside settlements of the same type, Colletière was a domanial fortified center. It sheltered a population of about 60 to 100 persons and was the result of an agrarian colonization which began from a near region (Rhône valley?) and took advantage of a population explosion around the Year On Thousand. Its inhabitants were expelled, c. 1033-1035, during a climate deterioration, sufficient for a water level rising, which again covered the site. It is at this time - unless the movement had begun c. 1025-1030- that the fortified motte of Châtelard (Chirens) was erected, at only one kilometre of distance, and many other ones in the near vicinity. These mottes construction, in the region like in Europe, mark, in a well-known climate of violence, the instauration of multiple private castellan powers, more often the adulterine core of the feudo-vassalic system to come.
These first attempts, real birth of the "feudal" Middle Ages, will not reach the critical height - geographic, human and economic - to succeed. Thus, they will rapidly leave their place to a more restricted number of fortifications, which will give birth to the castellan seigneury of Clermont, Virieu and Paladru. In their turn, they will be included progressively in the powerful barony of Clermont, which is located in the limits of Dauphiné and Savoie, between Guiers and Isère.
Still, numerous questions are to be. answered : was there at Colletière a dependant settlement ? Can we define both the morphology of the fortification and its entrance ? Serious hypothesis exist about the location of the contemporaneous cimetery. Did it have, as we presume, a church ? How was the private fortification's morphological aspect which remains, during the passage between the 11th-12th centuries, before the castles construction of which we can study today their remnants (from the 13th century) ? What did the peasant's settlement in the middle of the 11th century, at the 12th and 13th centuries (part of the incastellamento in the soil's structuration) look like ? And of course, through each interpretation lies the juridical aspects: Colletière's inhabitants' status, importance of the allodium, which certainly was important in Sermorens and in Dauphiné.
But, if we can make many comparisons between Paladru's lake region and its evolution at the 10th-12th centuries, and other sites which were studied elsewhere in France or in Europe, and if finally Colletière seems to be, after 20 years of study, representative of the quick mutation of this time, its major interest, thanks to the lacustrian surroundings, is to offer information at the crossing of technical analysis and various methods of study, of History and "traditional" Archaeology to the paleoenvironmental approaches. Colletière's study illustrates, specially in that period when technology had been even less evolved for almost ten centuries, the extremely interdepending worlds of men and nature. And finally, this particularly characteristic example shows us how quick and decisive the" Year One Thousand Revolution" was.
