Art by firelight? Using experimental and digital techniques to explore Magdalenian engraved plaquette use at Montastruc (France)

Abstract

Palaeolithic stone plaquettes are a type of mobiliary art featuring engravings and recovered primarily from Magdalenian sites, where they can number from single finds to several thousand examples. Where context is available, they demonstrate complex traces of use, including surface refreshing, heating, and fragmentation. However, for plaquettes with limited or no archaeological context, research tends to gravitate toward their engraved surfaces. This paper focuses on 50 limestone plaquettes excavated by Peccadeau de l'Isle from Montastruc, a Magdalenian rockshelter site in southern France with limited archaeological context; a feature common to many art bearing sites excavated across the 19th and early 20th Centuries. Plaquette use at Montastruc was explored via a programme of microscopy, 3D modelling, colour enhancement using DStretch©, virtual reality (VR) modelling, and experimental archaeology, the latter focusing on limestone heating related to different functional and non-functional uses. While the limited archaeological context available ensures the results remain only indicative, the data generated suggests plaquettes from Montastruc were likely positioned in proximity to hearths during low ambient light conditions. The interaction of engraved stone and roving fire light made engraved forms appear dynamic and alive, suggesting this may have been important in their use. Human neurology is particularly attuned to interpreting shifting light and shadow as movement and identifying visually familiar forms in such varying light conditions through mechanisms such as pareidolic experience. This interpretation encourages a consideration of the possible conceptual connections between art made and experienced in similar circumstances, such as parietal art in dark cave environments. The toolset used to investigate the Montastruc assemblage may have application to other collections of plaquettes, particularly those with limited associated context.

Fig 1

Photographs and digital tracings of plaquettes 675 (a; c) and 667 (b; d) from Montastruc. Scale bar below each plaquette is 10cm in length.

Fig 2

DStretch manipulated images in the LAB colourspace showing the difference between DStretch images of replica limestone plaquettes in experiment E (A) before heating and (B) after heating. Note that some reddish hues are visible in the DStretch image of replicas before heating, that may be caused by adhering sediment or the natural colouration of the limestone. Thus, the use of DStretch alone is not reliable to identify rubefaction traces.

Fig 3. High-resolution photographs and DStretch© images of example replica plaquettes from each experiment.

(A) shows plaquettes from the taphonomy experiment that were in direct context with the fire, placed on the surface both uncovered (SU1: left) and covered with sediment (SC1: right). (B) shows replica plaquettes PB3 (left) and PB4 (right) from the boiling stone experiment. (C) shows replica plaquettes A3 (left) and D1 (right) from the hearth oven structure experiment, (D) shows replica plaquettes WP1 (left) and WP3 (right) from the water pour experiment. (E) shows replica plaquettes FS1.1 (left) and FS1.2 (right) from the fireside night experiment. See the S1 File for a full description of the results from each experiment. Scale bar below each replica plaquette is 8cm in length.

Fig 4

Photographs showing a comparison between Montastruc plaquette (a, b) and replica plaquette FS1.2 (c, d) used in experiment E. b and d are DStretch© manipulated images using the LAB colourspace matrix to enhance pink discoloration caused by heating. Scale bar below each plaquette is 10cm in length.

Fig 5. Photograph showing ambient light levels and the position of replica plaquettes in relation to the fire during experiment E.

Fig 6

High resolution screen captures of the 3D models of plaquette 691 (a, b) and plaquette 684 (c, d) from Montastruc in the VR simulation at the maximum and minimum illumination as the light flickered. For both plaquettes, aspects become more or less visible as the light shifts, adding to a sense of movement.

Fig 7

High resolution screen captures of the 3D models of plaquettes 662 (a, b) and 685 (c, d) from Montastruc in the VR lighting simulation. The different position of the viewer or orientation of the plaquette renders visible or ambiguates different figurative depictions on the plaquette. For plaquette 662 one orientation reveals a large ibex (a), and another reveals two ibexes facing each other (b). For plaquette 685, several horse heads can be perceived in one orientation (c) whilst a grazing horse can be seen in another (d).