British Museum blog

Changing faces: revealing ancient alterations in Saharan rock art

relevant image alt text

Jorge De Torres, Cataloguer, African rock art image project

The Ennedi Plateau cliffs

The Ennedi Plateau cliffs, Chad

Fifteen years ago, I started my training as an archaeologist participating in a rock art survey in Extremadura, Spain. For a month I climbed cliffs and endured summer temperatures of 45ºC, looking for the flat rock faces where the schematic rock art we were looking for might be. One of those exhausting days, I crawled under a shelter during a break to escape the scorching sun. It was so small that you could only lie down and it had no space to turn sideways or sit. I rested for a while enjoying the shade, and then I saw them: four vertical, red lines painted on the inner part of the roof, clearly the imprints of four human fingers, made by someone who was once in my exact position, in a place where nobody but he (or she) – and thousands of years later, me – could contemplate them.

I’ve seen quite a lot of rock art since that summer morning, but I’ve always recalled that painting as one of the most important archaeological remains I’ve ever come across. Not because of its complexity, of course, but because of the exceptional possibility of recording and understanding the concrete action of an individual who existed thousands of years ago. Archaeologists like me are used to focussing on tendencies (chronologies, styles, geographical distributions) rather than individual human actions, which are usually very difficult to detect. However, while cataloguing the incredible collection of the African rock art image project, I found two such cases – both attempts to amend a picture once it was painted.

Figure 2 Detail of the engraved women at Niola Doa

Detail of the engraved women at Niola Doa, Chad

The depictions are found in the Ennedi Plateau in the north-eastern corner of Chad, a mountainous region on the southern edge of the Sahara Desert, full of huge outcrops and boulders, many of them covered with engravings and paintings dated from 5000 BC onwards. Although rock art in the Ennedi Plateau has a great variety of styles and depictions, probably the best known images in the area are the Niola Doa engravings: several groups of large figures (probably women), richly decorated, with one arm stretched downwards and the other bent upwards, usually resting sticks on their shoulders.

While describing these images, one caught my attention: an elegant, richly decorated woman, painted in red and white. There are several white lines around the neck, representing necklaces, and several more around the waist and hips, including series of white dots – possibly objects sewn onto a belt or directly to the skirt, reminiscent of the coin and shell belts often worn by dancers in the Middle East.

The painted woman from Niola Doa, before and after digital enhancement, showing the repainted arm position.

The painted woman from Niola Doa, before and after digital enhancement, showing the repainted arm position. Click on the image to see a larger version of the original.

Detail of the corrected arm from the painted woman

Detail of the arm from the painted woman, after colour enhancement.

There was something strange about the figure’s left arm: a red band under the left elbow, undoubtedly something painted, but a bit out of place. Using colour enhancement tools, such as those described by Elizabeth Galvin in a previous post, the result was astonishing. The enhanced photograph shows how the lower stain is in fact an arm that was painted stretching downwards, later corrected and repainted to bend upwards. The earlier arm is faint, but the enhanced colour shows how the tonality of both paintings is the same, implying that the same painter corrected the figure. Why was the image changed? We can only guess, but the final outline of the woman resembles the engraved figures of previous periods, so perhaps the painter was trying to emulate the impressive engravings that still give the place its name today (Niola Doa means ‘the dancing maidens’ in the local language).

The Archei Geulta (water pocket), Chad

The Archei Geulta, Chad

The second example comes from a very special place known as the Archei Guelta. A ‘guelta’ is a pocket of water in the desert (sometimes an oasis, but not always) that provides vital water to both people and animals. The Archei Guelta is one of the most important places in the region, with water available all year round, and home to one of the last remaining colonies of crocodiles in the desert. Like many other areas of the Ennedi Plateau, the whole area is full of paintings and engravings of many different periods and styles.

Painted panel of riders in ‘flying gallop’ style, before and after digital enhancement. The horse at the top is superimposed on an earlier painting of a man. See a larger version of the original image.

Painted panel of riders in ‘flying gallop’ style, before and after digital enhancement. Click on the image to see a larger version of the original.

One of these paintings is an extremely faint group of riders on horses, depicted in a very specific style of the Ennedi Plateau known as the ‘flying gallop’. Being so faint, the images were difficult to describe, and therefore I again had to use colour enhancement to identify them. By inverting the colours, I was able to see the riders and some previously undetected cows , but it also led me to an unexpected discovery: one of the riders was in fact a man on foot, with a horse superimposed. The paint of the man was much more degraded than that of the horse, implying that he was painted in an earlier period, perhaps prior to the introduction of horses to the desert. As in the first case, we can only speculate as to why the painters of the horses decided to amend the figure, but perhaps it was a way of incorporating older figures into the new scenes, adding as prestigious an animal as a horse. Perhaps they simply felt sorry for the lonely man walking among fast, powerful riders.

Detail of walking man superimposed by horse

Detail of the enhanced image showing a galloping horse painted over a standing man.

These two examples remind us that behind the broad categories into which we organize rock art were individuals who used these wonderful paintings and engravings as a way of sharing their own perspectives and interpretations of reality. The reinterpretation of older images raises interesting questions about how these populations interacted with their own past, integrating it within their narratives. And although the ultimate meaning of these changes can be difficult to comprehend, they nonetheless help us feel nearer to the people who made these images so many thousands of years ago.

This post is part of the African rock art image project at the British Museum, generously supported by the Arcadia Fund.

Filed under: African rock art, Archaeology, Collection, , , , , , , , ,

8 Responses - Comments are closed.

  1. Alex travois says:

    I don’t believe the repainting of the arm was a mistake. It looks like she is holding a glass in her hand and the other arm is showing the motion of which she is in the act of consuming something.


    • Blog author Jorge de Torres:

      First of all, thank you for your interest in this post and the project, as well as your comments about the first example.
      What you describe (figures represented with multiple limbs to depict movement) is not seen in the Sahara. The only known images in the Sahara that appear to show multiple limbs are some representations of horses in profile, with legs outstretched as if running (see BM 2013,2034.4567). However, this has been widely interpreted as showing depth perspective of multiple horses pulling a chariot, rather than one horse with many legs depicted to show its movement.
      In the rock art of this region and period, motion is not depicted by the representation of the same limb in different positions. When movement is depicted, it is often done in an ‘action shot’, for example by legs extended (as in the case of horses galloping or men running), or crossed (in the case of camels, to imitate their peculiar movement). Most human images of this period are in fact very static: facing forwards and typically with arms stretched downwards. The exceptions are running warrior figures holding weapons, riders, or people performing specific actions such as playing instruments. None of these cases has movement depicted with multiple arms.
      Regarding what the figure is holding, we can only speculate as to what it is.


  2. I love how ancient artists second guessed themselves and painted over things (the repainting of the arm). The more people change, the more they stay the same. I bet he or she was really annoyed that you could still sort of see the straight arm going down. “Oh, great. Now she has TWO arms!”


  3. Anne Stoll says:

    Have used my enhancement program (DStretch) on a captured jpeg of your lady and agree with Alex, she seems to be holding a cup or something with the same white paint in the center as the lovely white beads/shells around her hips, etc. I would suggest you have here simple superpositioning — perhaps three episodes — with really no way to deduce visually how much time has elapsed between them. A nice story about a single artist’s vision, but don’t see the support for this. BTW, have photographed many images in Namibia, SA, and Zimbabwe with white painted highlights applied on top of a red/brown base figure, assumed to have been done much later but again, no proof.


    • Blog author Jorge de Torres:

      Thank you for your comment and your interest in the project. Throughout Chad, there are many depictions similar to this figure, but with both arms stretched downwards. Since the tone of dark paint in the modified arm is nearly identical to the rest of the body, this points to a correction very soon after the original painting, although it doesn’t prove it conclusively (this is impossible without scientific testing). The evidence therefore points to the original image as having been conceived with both arms pointing downwards and then modified within a very short time. This could have been done by the same person or by somebody else – we cannot know.
      As you state, we also cannot know exactly when the white pigment was added; it may have been in three separate sessions, or at the same time as the other work.

      Thank you again for sharing your thoughts.


  4. Sam Williamson says:

    The flying gallop example is beautiful, it’s staggering how much detail there is just waiting to be uncovered, what fantastic work!


  5. vicki s says:

    there were horses in ancient Africa?


  6. Anne Stoll says:

    Now there’s a good point. That would affect the date estimates for the art, surely. When did horses and horse-back riding arrive in this part of Africa? Any guess?


Enter your email address to follow this blog and receive notifications of new posts by email.

Join 16,401 other followers


Follow @britishmuseum on Twitter

%d bloggers like this: