Navigation – Plan du site

AccueilNuméros60Amidst the waves, the Roc’h Sante...

Amidst the waves, the Roc’h Santeg Leton rock shelter (Santec, Finistère)

An urgent excavation of Middle Palaeolithic to Iron Age occupations in stratigraphic position
Grégor Marchand, Marie-Yvane Daire, Chlöe Martin, Caroline Mougne, Anne-Lyse Ravon, Pau Olmos Benlloch, Caroline Hamon et Marylise Onfray
avec la collaboration de Yann Bernard, Laurent Quesnel et Daniel Roué
Traduction de Louise Byrne
p. 67-69
Cet article est une traduction de :
Au milieu des flots, l’abri-sous-roche de Roc’h Santeg Leton (Santec, Finistère) [fr]

Résumés

La fouille d’urgence menée en 2015 et 2016 dans l’abri-sous-roche de Roc’h Santeg Leton (Santec, Finistère) est née de réflexions convergentes sur les abris-sous-roche (programme « Tous aux abris ! ») et sur l’érosion du patrimoine littoral (programme « ALeRT »). Le site est installé sur un îlot rocheux à 1 300 m de la côte et a été découvert par D. Roué en 1985. Neuf moments d’occupation ont été enregistrés dans une stratigraphie de 1,10 m d’épaisseur. Relativement modique, le mobilier archéologique recueilli lors de la fouille comprend des tessons, des charbons, des rares coquilles de noisette et des pièces lithiques (silex, quartz, macro-outils en granite). Le Paléolithique moyen est attribuable à une phase récente, soit pendant le MIS 5 (5d/5a), soit durant le Weichselien récent (MIS 4 à 3). Le mobilier lithique mésolithique est rapportable au groupe de Bertheaume (Premier Mésolithique) et au Téviecien de faciès Beg-er-Vil (Second Mésolithique). En sommet de stratigraphie, un petit foyer en cuvette est daté par le radiocarbone de la fin de la période gauloise (La Tène) et accompagné de tessons attribués au début/milieu du second âge du Fer. Une petite fosse contemporaine était remplie de coquilles, essentiellement des patelles. L’analyse géoarchéologique montre que le signal « littoral » n’est pas perçu à travers la dynamique pédo-sédimentaire qui est largement contrôlée par des apports colluviaux et des ruissellements concentrés. Cette opération de sauvetage a permis de valider des méthodes d’intervention rapide, dont n’est jamais exclu le tamisage à l’eau, seul moyen d’échantillonner correctement des sites préhistoriques.

Haut de page

Texte intégral

1The urgent excavation carried out on the Roc’h Santeg Leton islet (Santec, Finistère) came about as a result of convergent reflections on rock shelters (“Tous aux abris !” programme; Marchand and Naudinot 2015, Naudinot and Marchand 2020) and the erosion of coastal heritage (“ALeRT” programme; Daire et al. 2012, López-Romero et al. 2013). It is also part of ample work on the Palaeolithic in the region of Brittany, conducted under the leadership of J.-L. Monnier since the 1970s, which gave rise to settlement modelling highlighting the importance of coastlines (Monnier 1980, 2006, Monnier and Le Cloirec 1985, Monnier et al. 2016).

2This rocky block is part of the group of islands of the Santec coastline (Finistère) and is situated at 1,300 m from the coast; fig. 2). Access to the islet by foot is not easy and it can only be reached at low tide with a coefficient of at least 90, by crossing the rocky foreshore from île Verte (Enez Glaz). As a result of this difficult access, the site was not subjected to major anthropogenic pressure and the remains are in a good state of conservation; but this constraint also hampered access and the monitoring of the regular erosion of the archaeological remains by the different winter storm episodes. There is practically no doubt that this rock shelter was on the mainland during the Pleistocene and at the beginning of the Holocene (García-Artola et al. 2018). However, the insularity of the site during the Iron Age occupation, in the second half of the last millennium BCE, is still uncertain. It is difficult to formulate a definitive opinion without a more precise study of local bathymetric conditions, although we estimate the level of the sea to have been around 1 m lower than the present-day level.

Fig. 2 – Position on the foreshore of the Roc’h Santeg Leton islet (Santec, Finistère), in red on the image

Fig. 2 – Position on the foreshore of the Roc’h Santeg Leton islet (Santec, Finistère), in red on the image

The bathymetric curve of 0 asl (mid-tide) is shown in black. The bathymetric curve of the lowest spring tide is indicated in green

IGN topographic fund, map by H. Duval and G. Marchand

3The site was discovered and reported to the authorities by a voluntary archaeologist from Santec (D. Roué) in 1985. Only two short notes briefly presented the main objects discovered between 1985 and 2010. A first week of excavations in March 2015, under the leadership of P. Olmos Benlloch, established a stratigraphic outline, gathered archaeological objects and measured the overall topography of the site. The Azilian turned out to be totally eroded, but two Middle Palaeolithic levels were identified at the base of the 110 cm thick stratigraphy. A larger scale excavation for two weeks in July 2016 allowed for the exhaustive recording of the Holocene remains and a better evaluation of the Middle Palaeolithic remains.

4The site lies on a “platform” framed by two large rocks (South and West rocks) on the top of the islet, and is organized into four zones: Zone 1 corresponds to the in situ sediments, Zones 2 and 3 correspond to the stratigraphic sections outside Zone 1, and Zone 4 is the interior of the very leached rock shelter (fig. 9). Zone 1 is the main excavated area and measures 3 m wide and 11 m long (with only 8 m2 of preserved sediments, the rock outcrops to the southwest). The Holocene remains (Mesolithic and La Tène) are mainly concentrated in the northern part as the western part of Holocene levels underwent sedimentary truncation. Only the Gallic hearth survived this erosion, as well as the lower indurated layer of Pleniglacial silts. The archaeological objects collected during the excavation comprise sherds, charcoal, rare hazelnut shells and lithic objects (flint, quartz, macro-tools in granite). The total number of objects is rather low: in addition to the 176 lithic objects discovered by D. Roué during surface collection between 1985 and 2010, a further 73 objects were extracted in 2015 and 890 in 2016.

Fig. 9 – Site map, showing archaeological structures, areas and sections

Fig. 9 – Site map, showing archaeological structures, areas and sections

Altitudes are in metres ASL

Drawings and CAD G. Marchand

5The thickness of the sedimentary deposits in Zone 1 is slightly more than 1 m. The basal part could have been deposited more than 100,000 years ago during the Eemian, whereas the aeolian silts were deposited during the Pleniglacial, about 20,000 years ago. The Tardiglacial and Holocene occupations lie directly on these very fine silts. The human occupations were recorded as a result of the remobilization of these sediments. The results of the micromorphological analysis show that the upper part of the loess sequence (SU 2014 and 1013) is made up of clayey silts including sands derived from the erosion of granitic sands, which were remobilized by water. It is only from SU 1103 onwards that the first signs of the anthropization of the sediments appear with the formation of activity surfaces structured by trampling and the integration of aggregates of leached silts. In Zone 1, four successive stratigraphic units were excavated and sieved over a thickness of around 20 cm, with La Tène sherds for the former and flint from the Early and Late Mesolithic in the latter, unfortunately with no strict stratigraphic cut-off point. At this stage, we can confirm that the most threatened archaeological levels were excavated and recorded. There are no longer any in situ or reworked Holocene levels and a 40 to 50 cm layer of Pleistocene aeolian silts protects the Middle Palaeolithic levels which we left on site (from 10 to 20 m2). If no exceptional climatic events occur, these levels are thus protected and can wait for an archaeological team to organize exploratory operations in the near future.

6The presence of Mousterian lithic remains in Zone 2 in the highest SU (SU 2002) is still tenuous and requires further discussion after future excavations. On the other hand, lithic remains are very abundant in the two lower stratigraphic units (SU 2003 and 2007). This series comprises 274 lithic elements and presents rather strong typo-chronological characters, such as a marked use of the Levallois method, the absence of denticulates or macro-tools, the absence of bifacial tools and the use of flint and quartz in almost equivalent proportions (fig. 26). They point to an occupation dating from the recent phase of the Middle Palaeolithic, during MIS 5 (5d/5a), or the recent Weichselian (MIS 4 to 3). The coastal implantation of the site of Roc’h Santeg Leton would be more consistent with MIS 5, when sea levels were higher, as MIS 4 and 3 occupations are generally sited further inland, in particular on large sites of raw materials (Monnier 2006). But this is clearly not a decisive chronological argument.

Fig. 26 – Roc’h Santeg Leton, Middle Palaeolithic lithic industry

Fig. 26 – Roc’h Santeg Leton, Middle Palaeolithic lithic industry

1-2: Levallois flint flakes; 3: laminar flint flake; 4-5: laminar flint flakes; 6: scraper by inverse retouch on Levallois flint flake; 7: scraper on Levallois quartz flake

Drawings and CAD A.-L. Ravon

7Transitions to the Mesolithic are recorded in Zone 1 in SU 1106 and 1103, but also in the upper reworked SU. The transition between SU 1103 and 1102 is marked by a clear truncation followed by a massive backfill type input. This type of reworking could have had an effect on the redistribution and mixing of Mesolithic objects with more recent objects. A radiocarbon date indicates an age of 7600-7575 BCE for SU 1102. The formation dynamics of SU 1006 and 1103 indicate an increase in the anthropization of the soil and the temporary settlement of the site. These occupations are installed on colluvial inputs which record some evidence of the mechanical effects of trampling. A hollow hearth with no stone walls was recorded in the lower part of this Holocene level on colluvial sediments. The radiocarbon dating places the hearth at the beginning of the 5th millennium BCE, during the early Neolithic.

8The lithic assemblage, consisting of just 651 pieces, is attributable to the Bertheaume group (Early Mesolithic) and to the Beg-er-Vil facies (only symmetric trapezoids; Marchand 2014; fig. 32) of the Teviecian (Late Mesolithic). It is essential to mention five wedges (pièces esquillées), four of which come from SU 1101. This type of a posteriori tool results from the use of a flake as a splitting wedge, during which repeated percussion produces a quadrangular shape and multiple scars of vibrating removals. One of them is a quartz flake. These tools are almost absent during the regional Mesolithic, but common during the Neolithic. In Zone 4, in the rock shelter itself, a strip of an archaeological level contained 18 knapped flint blocks (SU 4002). This is a genuine anthropogenic accumulation, which could be assimilated to a “hoard” of coastal flint pebbles, most of which were in the initial production phase. The surface material includes an Azilian point, the only conclusive evidence of a final Palaeolithic occupation.

Fig. 32 – Roc’h Santeg Leton, lithic tools from area 1 (flint)

Fig. 32 – Roc’h Santeg Leton, lithic tools from area 1 (flint)

1: retouched flake; 2: monotroncature; 3: symmetrical trapezoid; 4: backed bladelet; 5: proximal fragment of backed bladelet; 6: drill on semi-cortical flake; 7-9: wedged pieces (no. 7 on a quartz crystal).

Drawings G. Marchand

9The two pebble walls visible at the top of the main zone (Zone 1) are very recent and are probably related to the use of the site as a hunting shelter. At the top of the stratigraphy, a small hollow hearth is radiocarbon dated to the end of the Gallic period (La Tène). It was dug into a level of very indurated loess silts, surrounded by rubified stones placed on their sides, and comprised a side vent. The 234 sherds from the upper part of the stratigraphy are attributable to the beginning/middle of the second Iron Age, whereas two dates obtained in the hollow hearth designate the last three centuries BCE. Six macro-tools or anthropogenic elements are linked to these occupations, consisting of very slightly transformed pebble tools in granite. They were used occasionally, mainly for grinding or percussion. A small pit in Zone 3 (SU 3002) was filled with shells. Five marine shellfish species were identified, including four gastropods and a bivalve. The vast majority of these were limpets, regardless of the type of quantification used (99.8 % of the NR, 97.4 % of the MNI and 99.6 % of the PR; fig. 41).

Fig. 41 – Marine mollusc species present in Area 3, US 3002

Fig. 41 – Marine mollusc species present in Area 3, US 3002

1: limpet Patella vulgata (L = 38 mm), 2: limpet Patella depressa (L = 32 mm), 3: monodon Phorcus lineatus (L = 25 mm), 4: common mussel Mytilus edulis (L = 9 mm), 5: Gibbula sp. (L = 4 mm). Number on the plate: common name, Latin name according to CLEMAM 2015; L: maximum length of the remaining shell in millimetres.

Photo C. Mougne

10Nine occupation periods have been identified at the Roc’h Santeg Leton islet, in a 1.10 m thick stratigraphy. These multiple cultural components show that this rock shelter attracted the interest of human populations at diverse periods of prehistory, when the site was a rock at some distance from the coastline. The Middle Palaeolithic occupation of the Roc’h Santeg Leton site is a typical Mousterian with predominant side scrapers and Levallois technology, like for many sites in the Armorican Massif (Cliquet and Monnier 1993). Part of the lithic industry found at Roc’h Santeg Leton is attributable to the Early Mesolithic dated to around 7600 BCE. The narrow backed bladelets and a point with two retouched edges are part of the Bertheaume group, for which the regional reference is Toul an Naouc’h in Plougoulm (Kayser et al. 1990). Another part – with in particular symmetric trapezoids – is attributed to the Beg-er-Vil facies of the Teviecian. This is one of the rare occurrences of the Teviecian on the northern coastline of Brittany.

11The “coastal” signal is not observed through pedo-sedimentary dynamics, which are mostly governed by colluvial inputs and concentrated runoff. During the Iron Age, when the rock was surrounded by waves (island or a peninsula?), the probably temporary occupation participated in a dense network of activities linked to the exploitation of the foreshore. The Breton coastline, in the broad sense of the term (present-day islands and shoreline), appears at that time as a zone of dense occupation with “promontory forts” on cliff promontories, “open” villages containing domestic structures associated, in certain cases, with salt production craft installations, isolated graves, associated with dwellings, or cemeteries or even places of worship (Daire et al. 2015). The Santec coast comprises a large foreshore dotted with islands propitious to fishing and human activities, which were preserved by Pleistocene aeolian deposits and Holocene dune formations. This is one of the best-known communes in terms of coastal sites, owing, in particular, to the prospection and inventory work carried out by J.-C. Le Goff and D. Roué since the 1970s. This rescue operation validated rapid intervention methods, including wet sieving, which is the only way of properly sampling prehistoric sites.

Haut de page

Bibliographie

Cliquet D., Monnier J.-L. 1993 Signification et évolution du Paléolithique moyen récent armoricain, Bulletin de la Société préhistorique française, 90 (4), p. 275-282.

Daire M.-Y., Le Bihan J.-P., Lorho T. 2015 – Les modes de l’occupation du littoral de la Bretagne continentale à l’âge du Fer. Une première approche, in Les Gaulois au fil de l’eau. Actes du 37e colloque international de l’AFEAF, Monpellier, 8-11 mai 2013, Bordeaux, Ausonius Éditions, vol. 1, p. 143-166.

Daire M.-Y., López-Romero E., Proust J. N., Regnauld H., Pian S., Shi B. 2012 – Coastal changes and cultural heritage: Towards an assessment of vulnerability through the Western France experience, Journal of Island and Coastal archaeology, 7, p. 168-182.

García-Artola A., Stéphan P., Cearreta A., Kopp R. E., Khan N. S., Horton B. P. 2018 Holocene sea-level database from the Atlantic coast of Europe, Quaternary Science Reviews, 196, p. 177-192.

Kayser O., Le Goff J.-C., Roué D. 1990 – Le site mésolithique de Toul-an-Naouc’h (Plougoulm, Finistère), Revue archéologique de l’Ouest, 7, p. 23-29.

López-Romero E., Daire M.-Y., Proust J.-N., Regnauld H., Pian S., Schaeffer E. 2013 Le projet ALeRT : une analyse de la vulnérabilité du patrimoine culturel côtier dans l’Ouest de la France, in Daire M.Y., Dupont C., Baudry A., Billard C., Large J.M., Lespez L., Normand E., Scarre C. (dir.), Anciens peuplements littoraux et relations homme/milieu sur les côtes de l’ Europe atlantique. Actes du colloque HOMER 2011, Vannes, 28 septembre-1er octobre, Oxford, Archaeopress (BAR Int. Ser. 2570), p. 127-136.

Marchand G. 2014Préhistoire atlantique. Fonctionnement et évolution des sociétés du Paléolithique au Néolithique, Arles, Errance, 520 p.

Marchand G., Naudinot N. 2015 Tous aux abris ! Les occupations du Paléolithique final et du Mésolithique dans les cavités naturelles du Massif armoricain, Bulletin de la Société préhistorique française, 112 (3), p. 517-542.

Monnier J.-L. 1980 Le Paléolithique de la Bretagne dans son cadre géologique, Rennes, Université de Rennes (Travaux du laboratoire d’Anthropologie), 607 p.

Monnier J.-L. 2006 – Les premiers peuplements de l’Ouest de la France. Cadre chronostratigraphique et paléoenvironnemental, Bulletin du musée préhistorique de Monaco, 46, p. 3-20.

Monnier J.-L., Le Cloirec R. 1985 – Le gisement paléolithique inférieur de La Pointe de Saint-Colomban, Carnac (Morbihan), Gallia Préhistoire, 28 (1), p. 7-36.

Monnier J.-L., Ravon A.-L., Hinguant S., Hallégouët B., Gaillard C., Laforge M. 2016 – Menez-Dregan 1 (Plouhinec, Finistère, France) : un site d’habitat du Paléolithique inférieur en grotte marine. Stratigraphie, structures de combustion, industries riches en galets aménagés, L’Anthropologie, 120, p. 237-262.

Naudinot N., Marchand G. 2020 Take shelter! Short-term occupations of the Late Palaeolithic and the Mesolithic in the French far West, in Picin A., Cascalheira J. (dir.), Short-term occupations in Paleolithic archaeology, definition and interpretation, Cham, Springer (Interdisciplinary Contribution to Archaeology), p. 121-146.

Haut de page

Table des illustrations

Titre Fig. 2 – Position on the foreshore of the Roc’h Santeg Leton islet (Santec, Finistère), in red on the image
Légende The bathymetric curve of 0 asl (mid-tide) is shown in black. The bathymetric curve of the lowest spring tide is indicated in green
Crédits IGN topographic fund, map by H. Duval and G. Marchand
URL http://journals.openedition.org/galliap/docannexe/image/1863/img-1.jpg
Fichier image/jpeg, 1,7M
Titre Fig. 9 – Site map, showing archaeological structures, areas and sections
Légende Altitudes are in metres ASL
Crédits Drawings and CAD G. Marchand
URL http://journals.openedition.org/galliap/docannexe/image/1863/img-2.jpg
Fichier image/jpeg, 555k
Titre Fig. 26 – Roc’h Santeg Leton, Middle Palaeolithic lithic industry
Légende 1-2: Levallois flint flakes; 3: laminar flint flake; 4-5: laminar flint flakes; 6: scraper by inverse retouch on Levallois flint flake; 7: scraper on Levallois quartz flake
URL http://journals.openedition.org/galliap/docannexe/image/1863/img-3.jpg
Fichier image/jpeg, 546k
Titre Fig. 32 – Roc’h Santeg Leton, lithic tools from area 1 (flint)
Légende 1: retouched flake; 2: monotroncature; 3: symmetrical trapezoid; 4: backed bladelet; 5: proximal fragment of backed bladelet; 6: drill on semi-cortical flake; 7-9: wedged pieces (no. 7 on a quartz crystal).
URL http://journals.openedition.org/galliap/docannexe/image/1863/img-4.jpg
Fichier image/jpeg, 293k
Titre Fig. 41 – Marine mollusc species present in Area 3, US 3002
Légende 1: limpet Patella vulgata (L = 38 mm), 2: limpet Patella depressa (L = 32 mm), 3: monodon Phorcus lineatus (L = 25 mm), 4: common mussel Mytilus edulis (L = 9 mm), 5: Gibbula sp. (L = 4 mm). Number on the plate: common name, Latin name according to CLEMAM 2015; L: maximum length of the remaining shell in millimetres.
URL http://journals.openedition.org/galliap/docannexe/image/1863/img-5.jpg
Fichier image/jpeg, 704k
Haut de page

Pour citer cet article

Référence papier

Grégor Marchand, Marie-Yvane Daire, Chlöe Martin, Caroline Mougne, Anne-Lyse Ravon, Pau Olmos Benlloch, Caroline Hamon et Marylise Onfray, « Amidst the waves, the Roc’h Santeg Leton rock shelter (Santec, Finistère) »Gallia Préhistoire, 60 | -1, 67-69.

Référence électronique

Grégor Marchand, Marie-Yvane Daire, Chlöe Martin, Caroline Mougne, Anne-Lyse Ravon, Pau Olmos Benlloch, Caroline Hamon et Marylise Onfray, « Amidst the waves, the Roc’h Santeg Leton rock shelter (Santec, Finistère) »Gallia Préhistoire [En ligne], 60 | 2020, mis en ligne le 31 août 2020, consulté le 05 décembre 2020. URL : http://journals.openedition.org/galliap/1863; DOI: https://doi.org/10.4000/galliap.1863

Haut de page

Auteurs

Grégor Marchand

UMR 6566 CReAAH Centre de recherche en Archéologie Archéosciences Histoire, Campus Beaulieu, 35000 Rennes — gregor.marchand@univ-rennes1.fr

Articles du même auteur

Marie-Yvane Daire

UMR 6566 CReAAH Centre de recherche en Archéologie Archéosciences Histoire, Campus Beaulieu, 35000 Rennes — marie-yvane.daire@univ-rennes1.fr

Chlöe Martin

UMR 6566 CReAAH Centre de recherche en Archéologie Archéosciences Histoire, Campus Beaulieu, 35000 Rennes — martin.chloe.26@gmail.com

Caroline Mougne

UMR 6566 CReAAH Centre de recherche en Archéologie Archéosciences Histoire, Campus Beaulieu, 35000 Rennes — caroline.mougne@gmail.com

Anne-Lyse Ravon

UMR 6566 CReAAH Centre de recherche en Archéologie Archéosciences Histoire, Campus Beaulieu, 35000 Rennes  — anne-lyse.ravon@univ-rennes1.fr

Pau Olmos Benlloch

Institut catalan d’Archéologie classique, Plaça d’en Rovellat, s/n, 43003 Tarragona — polmos@icac.cat

Caroline Hamon

UMR 8215 Trajectoires, Université Paris 1 Panthéon-Sorbonne, MSH Mondes, 21 allée de l’Université, 92000 Nanterre — caroline.hamon@cnrs.fr

Articles du même auteur

Marylise Onfray

UMR 8215 Trajectoires, Université Paris 1 Panthéon-Sorbonne, MSH Mondes, 21 allée de l’Université, 92000 Nanterre — marylise.onfray.@yahoo.fr

Haut de page

Collaborateurs

Yann Bernard

Virtual-Archeo, membre associé UMR 6566 CReAAH, Campus Beaulieu, 35000 Rennes — ybernard64@gmail.com

Laurent Quesnel

UMR 6566 CReAAH, Campus Beaulieu, 35000 Rennes — laurent.quesnel@univ-rennes1.fr

Daniel Roué

264 rue Plaine, 29250 Santec — daniel2.roue@orange.fr

Haut de page

Droits d’auteur

Gallia Préhistoire

Haut de page
  • Logo Ministère de la Culture
  • Logo Logo CNRS
  • Logo Maison des sciences de l'Homme Mondes
  • OpenEdition Journals
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search