Acknowledgements

I would like here to thank both the municipality of Saint-Jean-d’Arvey and the Regional Service of Archaeology Auvergne-Rhône-Alpes for having authorized the achievement of this photographic coverage.

1. Overview

Dataset location

1The data set is available at Zenodo repository with doi: 10.5281/zenodo.6483372

Publication date

22022/05/11

Context

3This dataset was collected in the scope of a post-doctoral research at Aix-Marseille Université from 2016 to 2018 on prehistoric schematic rock paintings from southern France and the western Alps. These schematic rock paintings, for the most part in rock shelters, sometimes in caves, are spread from the Iberian Peninsula to Italy. They are composed of recognizable elements like human figures, deer and caprinae, and of a great diversity of figures whose nature is today much more difficult to identify (ramiforms, grids, dots, U-shaped figures…). Such schematic rock paintings were first identified at the end of the 18th century in Spain and studied from 1910 by Breuil and Cabré (Breuil 1933-1935). In the Iberian Peninsula, where these paintings coexist with Levantine figures, the adjective “schematic” was used from the first discoveries to distinguish summarily executed figures, mainly reduced to their structure, from precise, descriptive and finely depicted Levantine compositions. This definition in opposition to the Levantine imagery probably masks different realities.

4In southern France, even if the first discovery of schematic paintings dates back to 1920 (Bellin 1979), their existence and antiquity were only accepted from the study by A. Glory (Glory 1941, Glory et al. 1948, Hameau 2002) performed after the publication of the Iberian paintings by Breuil. Then, from the 1980s, the systematic study performed by Ph. Hameau brought him to identify recurrent choices of figures or characteristics of the rock-shelters (southern exposure, red color of the rock surfaces, visible and/or overhanging location and presence of periodic runoff of water) suggesting a homogeneity of this graphic expression (Hameau 2002, 2005, 2006, 2009, 2011 for general overviews) which also spreads over the western Alps (Ayroles & Porte 1984, Seglie & Richiardi 1988, Seglie et al. 1988, Gambari 1992, 1994, Nisbet 1994, Arcà 1995, Gambari 1995, Arcà 1999, Crosa Lenz 2001, Gambari et al. 2001, Gambari 2007, Arcà & Fossati 2012, De Giuli & Priuli 2012, Fossati 2013, Defrasne & Bailly 2014, Walsh et al. 2016).

5The aim of this post-doctoral research was to question the nature, the structure and the chronology of the corpus of schematic rock paintings today composed of 146 sites in southern France and Italian Piedmont (fig. 1). Chronological milestones should be deduced from the environmental or archaeological context of the sites or from the imagery itself and are all mainly located from the 5th to the 3rd millennium BCE.

The site

6The Trou de la Féclaz site is located at 1140 m.a.s.l. in the south of the Bauges moutains in the inferior part of the urgonian limestone cliff of the Peney mount (1356 m.a.s.l.; fig. 2). It overlooks the Leysse plateau, the access to the heart of the Bauges moutains, the Chambery basin and the Savoy valley. The name of the site comes from the closeness of a steep and narrow passing in the cliff wall to the Féclaz plateau.

7The rockshelter, facing south-east (130°) and located on a 2,50 m wide rocky shelf and a hundred meters long, is shallow and constituted of a series of spalling rock strata (fig. 3). Paintings have been identified on both vertical and horizontal surfaces of two strata over a length of 5 m. The Trou de la Féclaz site is one of the two sites with schematic rock paintings in Savoy and one of the northernmost of this schematic corpus. A small cave is located at the end of the shelf, at 15 m of the painted rockshelter. The Trou de la Féclaz site differs from most of the others sites by the number and good preservation of the paintings and the diversity of the used coloring matters.

8These paintings were reported by a student from Chambéry in 1970 and rediscovered two years later by Mrs Hirsch and Mollard from Saint-Jean-d’Arvey. Rock paintings have been recorded and published by Mr Ayroles and Porte in 1972 and 1984 (Ayroles & Portes 1972, 1984, Rey 2016). They had never been restudied ever since. It consequently appeared necessary to record this site with the digital enhancement and 3D techniques we have today. This study allowed to precise the previous tracing and to identify new paintings (fig. 4). 264 figures and paintings remains have been identified in the rockshelter. This graphic corpus consists of simple marks, mainly painted on the vertical surfaces, as well as groups of dots, structured or not, which constitute one of the characteristics of this site, painted on the ceiling of the rock shelter. One of them, subdivided by tracings, has been interpreted as a map in comparaison with rock engravings known in other Alpine areas, for example around Mont Bégo in the Alpes Maritimes (Ayroles and Portes 1972). This interpretation should however be treated with caution. Other identifiable figures include star shape figures, cross shape figures and circles of dots, as well as a figure composed of orthogonal lines interpreted by the previous authors as a human figure (fig. 4B).

2. Methods

Steps and sampling strategy

9This 2D recording of the painted wall from the Trou de la Féclaz rockshelter has been performed with the administrative permission of both the Regional Service of Archaeology Auvergne-Rhône-Alpes and the Saint-Jean-d’Arvey municipality that is the owner of the site. This was achieved according to a strict methodology (Le Quellec et al. 2013). The camera used is a Canon EOS 6D. This camera has a full frame sensor of 20 million pixels. Associated with a fixed focal lens (EF 50 mm 1.4 USM), it guarantees the high quality of the pictures. The 50 mm lens has been chosen since it is the one more closely matching the natural view and it restitutes the perspective without important distortions. The manual mode, shutter priority mode and the automatic mode coupled with the Auto White balance have been used. Flash with the TTL mode has been used when necessary.

10A systematic photographic coverage of the rock surfaces was done without prejudicing if there is paintings or not. The camera lens should be perpendicular to the rock surface and images are captured from the top to the bottom of the wall in the case of vertical surfaces or from the bottom to the outside in the case of horizontal surfaces. Pictures are systematically taken from left to right. The shooting distance varies according to the site configuration. Overlapping areas are necessary between each picture. Photographs are collected in RAW format.

11This systematic coverage of the painted wall is completed with close-up photographs of figures. This allows a more precise recording of the main paintings and a more reliable analysis of the superpositions. Such close-up pictures are generally used for publication.

12Photos are then processed with Digital Photo Professional 4 (DPP4). The camera lens systematically generates distortions, even weak, that are consequently corrected. Once pictures are converted into the Adobe RGB profile, they are systematically processed and compressed into .jpeg files for enhancing with the plug-in DStretch® of the Image J© software in order to precise and/or reveal the faint painted figures. It’s important to note that .jpeg files automatically produced by the camera during the shooting have lower quality that the .jpeg files produces with DPP4 from the RAW files that’s why these latter are not used in the scope of this study.

Constraints

13Even if a strict methodology is required and respected as much as possible to produce this photographic coverage, the shooting is constrained by the presence of a guardrail in front of the cliff and by the distance between some horizontal painted surfaces and the ground of the rockshelter. Consequently, distance between the camera and the rock surface varies. In the same way, the narrowness of the rocky shelf prevents the frontal and general shooting of the rockshelter.

3. Dataset description

Object name

14Trou_de_la_Feclaz(73)_Pictures

Type of data

15Primary data

16Photographs, spreadsheets

Format of data

17.jpeg, .csv

Creation dates

1814/07/2017

Dataset Creators

19Claudia Defrasne, Aix-Marseille Université – LAMPEA (UMR 7269) – LabexMED

Language

20English

Other informations

21Pictures are named from their location in the rockshelter. Four aeras have been differentiated as illustrated by the following schema (fig. 6). Letters A and B are used to distinguish each of the painted strata and numbers are used to identify the painted surfaces (1 for vertical surfaces and 2 for horizontal surfaces of the strata). Then, pictures are ordered from left to right of the rockshelter, gathering photos constituting the coverage and details of the paintings.

4. Reuse potential and limits

22The publication of this dataset is primarily oriented towards rock art specialists. Even if a rendition as faithful as possible is always sought after, the digital tracings of rock imageries are interpretations of their author from the original pictures and their enhancement. Normally, digital tracings are published together with the photography they are taken from but limits on the number of figures does not always allow us to do that. In the same way, this cannot be done for all the rock images of a published site. Moreover, even it is the case, the reader cannot reproduce himself the enhancement from the published figures because of their low quality. Consequently, the publication of datasets of photographs gives the reader the option to do his own tracing and compare it to the published one or use it in his own research.

23In the present case study, the corpus of schematic rock paintings is known from the Iberian Peninsula to the Italian Piedmont. To be understood, this ideological phenomenon attesting the flow of ideas should consequently be studied at the European scale. The geographical area being too huge to allow the access to each site of the schematic corpus, it appears essential to share primary data so that researchers studying it could make comparisons from photographs of sites they cannot visit in the scope of their study. This should be done being aware that it will not replace the in situ study and understanding of the rock images.

5. Expert statement

Philippe Hameau

Laboratoire d’Anthropologie et de Psychologie Cliniques, Cognitives et Sociales (LAPCOS)
Philippe.HAMEAU@univ-cotedazur.fr

24The present work consists of setting up a photographic database concerning the shelter known as Trou de la Féclaz, marked with motifs belonging to schematic iconography, usually dated to the 5th-3rd millennium BC. Its primary aim is to offer the researcher the possibility of consulting photographs taken without any conceptual preconceptions and to be able to work on the entirety of the walls, painted or not, without having to be satisfied with the few illustrations that usually accompany an article, of variable quality and necessarily selected by the author(s). The history of the discoveries of the schematic iconography, the hypotheses of the coherence of the site and the presentation of the site itself are succinct but correct. A sentence listing the main identifiable motifs of the site, with reference to the presentations of Ayroles and Porte, would be appreciated, even if one can admit that it contravenes the neutrality of the presentation. The reference to the truly pioneering work of A. Glory et al. 1948 (Revue Préhistoire, t.X p.7-135) should be added. The protocol followed for the photographs is perfectly explicit and can be conceived as neutral as possible. The metadata are easily exploitable according to a logic and coherence mentioned in the presentation of the protocol. The significance of such a work would certainly be measured at the end of the resumption, even if experimental, of the metadata by a person working on schematic iconography but ignorant of the configuration of the site.

Hameau Ph., 2021, 'Expert Statement' in Defrasne C. (2021), Photographic coverage of the Trou de la Féclaz site (Saint-Jean-d’Arvey, Savoy, France) — data paper, Préhistoires méditerranéennes, 2021, 9.1. DOI: 10.34847/nkl.f7abt96j