Chariots in Saharan rock art: An aesthetic and cognitive review

Abstract

There are estimated to be around 1200 depictions of painted and engraved chariots in Saharan rock art, a number which has doubled in recent years. However, because of the nature of the environmental conditions in the Sahara, the chariot is regarded as having played a minor technological and cultural role, and archaeologists have been criticised for attributing too much importance to these images. This enquiry focuses on the aesthetic component of these depictions, addressing the variation in chariot representations. In reviewing the significance of these depictions beyond their technological significance, the aim is to consider how they functioned socially, culturally and cognitively within a Saharan context.

Keywords

Rock art chariots Garamantes symbol aesthetics cognition

Introduction

Saharan rock art provides insightful glimpses into the past lives of peoples who once inhabited this now desert region. One particular type of depiction, the chariot, has presented researchers with the challenge of situating this motif within an intelligible archaeological and rock art-related framework. Globally, the chariot has been one of the great empowering innovations of history. However, in a Saharan context, its function and importance remains a matter of debate not only as a mode of transport but also as a rock art representation. While attention has been given to the technological and constructional aspects of chariot imagery in history (Aiken Littauer et al., 2002), little has been said about the aesthetic component of these depictions. Yet, in this context, it seems a particularly germane approach as some depictions are very clearly chariots represented in profile, pulled by running horses with a standing charioteer holding a whip (see Figure 1); but others are highly schematic, devoid of animals and human figures and in some cases barely recognisable as a vehicle at all (see Figure 2). There appears to be no clear uniformity in the way chariots are represented; yet simultaneously, there also appears to be some visual correspondences between these types of depictions that need some investigation.

Figure 1.

Flying gallop style chariot (inset) as seen at bottom of this painted tableau. Two further flying gallop chariots on the far right and far left. Tarssed Jebest, Tassili n’Ajjer, Algeria. ©TARA/David Coulson and the Trustees of the British Museum (BM 2013,2034.4569).

Figure 2.

Collage of types of chariots categorised as schematic: (a) Oued Lar’ar, South Oran, Algeria (from Lhote, 1982; reproduced by de Torres); (b) Djebel Ben Ghnema, Libya (from Gauthier and Gauthier, 1999; reproduced by Anderson); (c) Essouk, Adrar des Iforas, Mali (from Lhote, 1982; reproduced by Anderson); (d) El Bayed, Mauritania (from Lhote, 1982; reproduced by Anderson); (e) Seroka, Niger (from Lhote, 1982; reproduced by Anderson); (f) Bir Igueni, Western Sahara (from Vernet, 1993; reproduced by Anderson); (g) Aoufilal, Taouz, Morocco (from Lhote, 1982; reproduced by Anderson).

This enquiry takes as its starting point a collection of images included in the British Museum’s African Rock Art Image project; a collection of 25,000 digital images acquired from the Trust for African Rock Art based in Kenya. One of the great benefits of working on such a project is being able to take a broad view of rock art from across a wide geographical area; observing similarities, differences and adjacencies in representational styles.

The collection that forms the basis of this examination comprises 540 images (see Table 1), some of which come from the British Museum collection but are augmented with previously published works from Libya (Camps, 1989; Gallinaro, 2013; Gauthier and Gauthier, 1999; Graziosi, 1942; Lhote, 1976; Mattingly et al., 2003; Mori, 1965; Muzzolini, 1995), Algeria (Camps, 1989, 1993; Camps and Gast, 1981; Gauthier and Gauthier, 1999; Lhote, 1976; Mori, 1965; Muzzolini, 1995), Morocco (Camps, 1993; Camps and Gast, 1981; Gauthier and Gauthier, 1999; Lorblanchet, 1992; Muzzolini, 1995), Mauritania (Lhote, 1976; Vernet, 1993), Niger (Camps and Gast, 1981; Gauthier and Gauthier, 1999; Lhote, 1976; Muzzolini, 1995), Mali (Lhote, 1976) and Western Sahara (Vernet, 1993).

Table 1.

Distribution of painted and engraved depictions in the sample.

	Painted	Engraved	Total
Libya	18 (29%)	44 (71%)	62
Algeria	81 (40%)	120 (60%)	201
Morocco	0	96 (100%)	96
Mauritania	0	125 (100%)	125
Niger	0	29 (100%)	29
Mali	0	6 (100%)	6
Western Sahara	0	21 (100%)	21
Total no.	99 (18%)	441 (82%)	540

The most recent estimation of the number of depictions of chariots across the Saharan landscape has increased from around 600–700 to now more than 1200 (Gauthier and Gauthier, 2011), although not all of these are available in published works and as such will not form part of this enquiry. This, then, is not an exhaustive study but serves to make some preliminary enquiries into potentially a much more comprehensive project.

Styles of chariots

There are numerous ways to categorise these sets of images based on a variety of criteria, but I have followed the current literature on the subject which classifies according to a stylistic typology, broadly defined as (i) flying gallop style and (ii) schematic.

Flying gallop

The flying gallop chariot (Figure 1) is portrayed drawn by one, two or three horses. The vast majority, 84 per cent in this study, has two horses, and all are portrayed galloping at full speed, with a single charioteer standing on a small platform, holding a whip and sometimes straining forward. The horses are always represented in profile, and the chariot is either depicted in profile view or from a flattened perspective. The distinguishing feature of the flying gallop category is the kinetic and/or performative nature of the depictions.

Their distribution (see Figure 3) is confined to Libya and Algeria and includes the whole Tassili region, although they are fewer in the Acacus mountains, and infrequent in the eastern Hoggar (Muzzolini, 2001: 621).

Figure 3.

Distribution map of Schematic and flying gallop style depictions (Courtesy. Jorge de Torres, British Museum).

It has been estimated that of the published 600–700 chariot images, a little more than a tenth of the total are depicted in this style (Allard-Huard, 1985: 35); just over 8 per cent of the collection analysed here is depicted in the flying gallop style. Within this category, chariots can be embedded in scenes of social engagement (see Figure 1), with figures standing, seated in groups, dancing and talking; pastoral scenes; groups of men running with sticks; and even depictions of oases with clusters of date palms. The chariot and charioteer are not necessarily the focus of the tableau but act as an integral element of a social, economic and cultural landscape in which desert peoples see themselves. In contrast, flying gallop chariots can also be represented in isolation or with few associated images so that they are explicitly the visual focus of the panel.

It has been suggested the ‘flying gallop’ style, as a representational convention, may have had local precedents in earlier rock art of the Tassili n’Ajjer, comparable to depictions of leaping gazelles and running hunters armed with sticks or bows (Allard-Huard, 1985: 47). This is a thought-provoking observation about how artistic practices accommodate and appropriate traditional conventions while introducing new motifs, an idea that I will return to later in the paper.

Schematic

While the flying gallop category is a well-defined group of depictions, the remaining imagery is less so, and is more problematic to characterise. What I will attempt to do here is to account for some of the different kinds of chariot styles, broadly describing variations and their distribution. Images of chariots vary from being reduced and simplified in form to more elaborate and detailed (see Figure 2); although the simplicity of form does not necessarily negate their capacity to be representational. Rather, they often show the relationship between the composite parts of the chariot.

In all chariot depictions in this category, wheels are shown in profile joined by an axle, many have spokes but some do not. The body of the chariot is depicted variously and can be perceived as being flattened (see Figure 2(c)) although some show a lattice-style body of the chariot in profile view (see Figure 2(a)) resembling the woven bodywork of racing chariots. In addition, there is variation in the presence, absence and quantities of shafts (single, double or multiple) and yokes. Moreover, some of these schematic chariots are depicted closely aligned and in numerous quantities (see Figure 2(g)). The ‘flattened’ perspective and the paring down of the basic features of the chariot to its constituent parts must be regarded as a significant visual convention, because it spans a vast geographical area.

Interestingly, except in Algeria, if animals are present they are more likely to be represented side-on and back to back (see Figure 2(e)) than in profile view (see Table 2). However, in some cases, this is not an overwhelming majority and in this analysis only demonstrates that both types of representations are practised. The flying gallop category always depicts horses in profile. However, in the schematic grouping, in some cases, animals may be defined by an engraved contour only (see Figure 2(c)).

Table 2.

Depictions of horses in profile and side on and back to back in the schematic category.

Horses/oxen depictions	Algeria	Libya	Niger	Morocco	Mauritania	Mali	Western Sahara
Side on and back to back	1	21	7	5	15	2	1
Profile	7	8	1	3	10	0	0

However, the vast majority, nearly 69 per cent of examples in this analysis, are depicted without any animals (see Table 3).

Table 3.

Number of chariot depictions absent of horses/oxen.

	Algeria	Libya	Niger	Morocco	Mauritania	Mali	Western Sahara
Absence of horses/oxen	119 of 201	19 of 62	21 of 29	88 of 96	100 of 125	4 of 6	20 of 21

The frequent absence of horses and oxen in the schematic category is a notable visual aesthetic, particularly for a vehicle that is only functionally operative when attached to its tractive force. This is made more prominent when compared to the flying gallop style of depiction, whose visual power lies in the explicit portrayal of speed and motion.

The vast majority of the painted and engraved rock art images of chariots found from the Fazzan in Libya through the Aïr mountains of Niger into northern Mali and then westward to the Atlantic coast incorporating Western Sahara, Mauritania and Morocco are depicted in a variety of schematic styles, where charioteers and often horses or oxen are customarily absent (see Figure 3). In addition, some of the most schematic in appearance are from the furthest west of the region.

The variation in detail and depictions, as well as the extent of their distribution, makes this category problematic to classify as simply ‘schematic’, because it does not seem to convey the complexity of the grouping. It is not within the scope of this paper, but a thorough and systematic analysis of this grouping is required to highlight this interesting detail and variability. Further, more detailed enquiries may provide some insight into regional specificities which correspond to particular cultural groups, trade routes or related rock art styles.

Stationary and equine

In addition to the flying gallop and schematic categories is a grouping of images, that I have termed here as ‘stationary and equine’. These are not classified as such in the scholarly literature, but they are noticeable within this survey and require some explanation. This collection includes 34 images from Algeria, Libya, Mauritania, Morocco and Niger, comprising depictions of chariots (either in profile or flattened view) pulled by horses or oxen that are either walking or appear stationary, as well as chariots that are unharnessed but in association with horses/oxen and equine/bovine attendants (see Figure 4). The attendants to the horses or oxen are depicted either standing in front of the animal with arms outstretched towards the muzzle, walking beside the animal holding the reins and a raised stick, holding reins as if in readiness to put them on the animal or leading it by the reins. In four images from Algeria and Libya, it appears as if the chariot has been uncoupled and upended as if ready to be stored.

Figure 4.

(Right) Upended chariot, two horses and attendants. Wadi Aramat, Tassili n’Ajjer, Libya. © TARA/David Coulson and the Trustees of the British Museum (BM 2013,2034.2099).

The presence of charioteers, both standing and seated, as well as passengers seems to be a feature that predominates in Algeria and Libya (see Table 4), becoming less evident further south and west. This category demonstrates the use of chariots, not necessarily in motion such as the flying gallop, but being pulled by animals, with charioteers, and in some cases with passengers, showing their potential social use. In no cases do the chariots appear to be carrying any loads, implying their use as vehicles of trade and exchange, but potentially this function may be implicit in particular types of representation, or their placement in the landscape.

Table 4.

Number of chariots in this category with standing or seated charioteers as well as those with passengers.

Country	Charioteer standing	Charioteer seated	Passengers
Algeria	23	9	5
Libya	10	2	2
Niger	4	0	2
Morocco	2	0	0
Mauritania	3	0	0
Mali	0	0	0
Western Sahara	0	0	0

These rock art images go some way to demonstrating the social and cultural milieu of chariots. A chariot is a sophisticated piece of craftsmanship, but one in need of ‘watchful maintenance when in use and careful preservation over periods of inaction’ (Piggott, 1992: 43). When not in use, the wheels of chariots had to be removed to avoid distortion and the whole vehicle leant against a wall. Chariots required a significant investment of time for preservation and conservation purposes. Moreover, a ready supply of horses had to be available, along with special training, pasturage, stabling and trained equine staff. Indeed, the ‘domesticated, broken-in, trained horse was in itself as much of an artefact as the chariot’ (Piggott, 1992: 42). Both the stationary and equine and flying gallop groupings of chariot depictions demonstrate in varying degrees not simply the adoption of a new method of transportation but the adoption of a package deal, what Piggott calls a techno-complex (Piggott, 1992: 45).

Distribution

The distribution of chariots across the Saharan landscape needs some further clarification. Rock art depictions, by necessity, are concentrated in or dispersed around rocky outcrops or mountain ranges where engravings or paintings could be made (see Figure 3). This may be an obvious point to make, but the distribution maps need to be understood in this context, so that many of the blank regions correspond to the vast expanses of sand sea where neither paintings nor engravings were possible. On what grounds particular spaces or places were selected to make chariot depictions is debatable; they might have been chosen because of their proximity to known routes used for trade or pastoralism. Particular sites may have come to signify a trading post or an area where middlemen might be willing to trade and/or barter, or even where chariots were made by skilled craftsmen.

The absence of archaeological evidence for chariots along the routes where their depictions occur has led to the suggestion that some representations of chariots may have been the result of cultural diffusion, transmitted orally by nomadic peoples traversing the region; artists may never have actually seen the vehicles themselves (Coulson and Campbell, 2010: 160). This is an interesting proposition and not unknown in other contexts, but determining which depictions might be the result of oral transmission rather than visual experience is almost impossible to identify.

However we think about chariot imagery, it is influenced by our understanding of their temporal and spatial relationships. The dating of rock art is inherently problematic and the chronologies attributed to the appearance of chariots in the Sahara remain contested.

Chronologies

The chariot is generally thought to have been introduced by the Garamantes, a cultural group descended from Berbers and Saharan pastoralists, who settled in central Libya from around 1500 BC (Mattingly et al., 2003: 287). The flying gallop chariot is the type that is most associated with the Garamantes. There is little textual information about them; documentation comes mainly from Greek (Herodotus, IV; Strabo, XVII ch. 3; Pliny the Elder, V ch. 5) and Roman (Tacitus, Annals III and IV; Histories IV; Ptolemaeus, IV) sources. However, recent excavations in the Fazzan by British (Mattingly et al., 2003, 2007, 2010) and Italian (Liverani, 2000) teams of archaeologists have contributed significantly to our understanding of this period.

The Garamantian state was centred on the area now known as Fazzan, the south-western region of modern Libya, but their influence was felt as far as northern Libya and the desert regions south of the Roman province of ‘Africa Proconsularis’ (present-day northern and eastern Tunisia, the Mediterranean coast of present-day western Libya and northeast of present-day Algeria) (Wilson, 2005: 224). The peak of Garamantian success was between 300 BC to AD 500, but their overall rise and decline spans 1000 BC to AD 500 (Mattingly and Sterry, 2013: 504). The Garamantian kingdom developed into what has been described as, ‘a true and proper state, with monumental architecture; something that in Fazzan had never happened before, nor was to happen again (on a comparable scale) probably until early modern times’ (Liverani, 2000: 18).

Trade networks engaged in by the Garamantes are clear from the archaeological record. Locally sourced carnelian beads were traded with the Romans (Wilson, 2005: 416), while Garamantian tombs of the first and second centuries AD contain Gaulish, Roman and Egyptian pottery and glass (Keech McIntosh, 1996: 748) as well as fig and pomegranate seeds and olive pits – all fruits imported from the coast (Werner, 2004). The Garamantes controlled trans-Saharan trade routes, and it has been estimated that the region under their influence covered an area of approximately 100,000 square miles (Mattingly, 2005).

The Greek historian Herodotus (IV, 183) described the Garamantes as using four-horse chariots to hunt ‘troglodyte Ethiopians’. Within the collection analysed here, there are only 12 instances where depictions of chariots show four horses; three from Algeria, five from Libya and four from Morocco. Of these 12, only two from Algeria are depicted in a flying gallop style characteristic of the Garamantes, but are not involved in a hunting scene. The remaining 10 four-horse chariots all show horses and chariots in a flattened style and convey a very static, fixed attitude. Mattingly et al. (2003: 88) has suggested that the intended human prey may have painted many of the flying gallop chariots from the Acacus and Tassili mountains. This is an innovative suggestion, but flying gallop chariots are more likely to be depicted amid scenes of social engagement, rather than conflict. Moreover, it is more likely that flying gallop images were created by chariot-using groups whose intention was to convey the ownership of such a vehicle and their skill in chariot racing.

French archaeologist and ethnographer Henri Lhote (1982) proposes that the structure of the four-horse chariots found in the rock art of the Fazzan, which corresponds to those described by Herodotus (c. 440 BC), act as a chronological marker for these types of chariots (Lhote, 1982: 28). He suggests that the flying gallop chariots were produced several centuries prior to Herodotus’ texts (1982: 28), although he does not propose any dates, and that they show a spatial progression towards the south of the region. The Adrar des Iforas (a massif in Mali that extends into Niger) and the Aïr mountains (Niger) show engraved chariots in what Lhote proposes is a later style, gradually replaced by riders in the Libyan-Warrior style and in some cases accompanied by Libyco-Berber script (Lhote, 1982: 30).

Lhote acknowledges the stylistic variation of engraved schematic chariots of central Sahara, Western Sahara and south of Algiers, and proposes that these are undeniably much later than the flying gallop (without specifying how much later) but are not all necessarily contemporaneous (Lhote, 1982: 28). In Tassili, for example, he observes that patinas on engraved images are sufficiently different to suggest that schematic images have been made at different times (Lhote, 1982: 28).

Not only does Lhote see the styles of chariot depictions as following a spatial and temporal sequence, but they correspond to where one finds today different cultural groups such as the Tebous (north of Chad around the Tibesti mountains), the Tuareg (predominantly Niger, Mali and Algeria), and Maures (or Moors of Morocco and Algeria), and suggests that this situation must have been in existence towards the start of the Christian era. In respect to the central Sahara, the Garamantes, mentioned by Herodotus, are thought to be ancestral Tuareg (Lhote, 1982: 30). The Tibesti mountains, Lhote suggests, can be identified as the country of the Ethiopian troglodytes as discussed by Herodotus, and the Aïr mountains as a region subservient to the Garamantes (Lhote, 1982: 30).

In summary, Lhote supposes that the introduction of the chariot to the region is several centuries prior to Herodotus’ reference to four horse chariots c. 440 BC, which may correspond to around the 8th or 9th century BC. He argues for a spatial and temporal artistic progression, from the central Sahara towards the south, with the western region of the Sahara being stylistically and chronologically separate and later than the flying gallop, and showing much artistic variation.

In contrast, the Italian prehistorian Muzzolini (1992) argues that the view that schematic chariots are more recent than the flying gallop seems to have developed from a preconceived notion that there are no examples of a schematic chariot superimposed on any flying gallop depictions. However, he suggests that this is because stylistically, painted (flying gallop) and engraved (schematic) chariots belong to two different zones. Aesthetic differences between chariot depictions are, for Muzzolini, ‘two different expressions of the same reality’ (1992: 15) created by cultural groups which are geographically distant, but that some of the chariots of each group are ‘doubtless contemporaneous’ (Muzzolini, 1992: 15). In Fazzan, schematic chariots are technologically similar to the flying gallop style and, as such, Muzzolini sees no reason to dissociate the two; all chariot depictions, engraved or painted, schematic or flying gallop can be taken as being relatively contemporaneous. The representational differences we see are the result of different artistic manifestations.

Muzzolini critiques Lhote’s dating of schematic chariots and calls for its abandonment based on his use of radiocarbon dating methods. Lhote proposed that chariot depictions from Bled Itini in Mauritania indicate an average date of 515 BC which is also used as a mean age for the chariots in the Aïr mountains of Niger because they are of the same schematic type (Muzzolini, 1992: 15). Muzzolini’s objections are based on the fact that the dates are taken from a nearby dwelling site, not the engraving itself, and the link between the engraving and site is not certain (Muzzolini, 1992: 16).

Muzzolini proposes that chariots were introduced to the Sahara from Cyrene, a city founded in 630 BC by the Greeks on the Libyan coast and that the flying gallop chariots imitated and conveyed the fame of their prestigious chariot races (1992: 28). Shortly afterwards, some horse-drawn chariots appear in the eastern Aïr mountains and Djado region of Niger in the later phases of the Libyan Warrior period (1000 BC–AD 500). Chronologically, this corresponds to the Horse Period (1000 BC–AD 1) in the Tassili, and during this period, the fashion for chariots spread through Fazzan, Tassili-Acacus, Hoggar and Maghreb (up to and including Morocco). Interestingly, he does not identify the Garamantes as the cultural group responsible for introducing the chariot from Cyrene.

A link with the Cyrenean races can be put forward for the flying gallop chariots only, explicitly expressing a particular vogue and performance of chariot racing which disappears before the advent of the Camel Period (1st millennium AD). Muzzolini agrees with Lhote that the schematic chariots in the west of the region may be slightly more recent than the flying gallop; however, the remainder date to the same period (Muzzolini, 1992: 17). Any stylistic differences in chariot representations, he proposes, are simply different pictorial strategies practised by cultural groups who are spatially discrete.

Researchers are unable to agree on the dates the chariot was introduced to the central Sahara, and whether the flying gallop and schematic categories are contemporaneous or follow a spatial and temporal trajectory. Problems arise because the schematic type of chariot is not well-defined, its distribution is spatially extensive and, stylistically, the flattened style of chariot depiction that has been classified as schematic is also evident in the flying gallop category. Thus, any aesthetic arguments rest on chronologies and until we have some concrete archaeological evidence and/or scientific dating, then the stylistic question remains problematic. Nevertheless, the aesthetic variations that both Lhote and Muzzolini acknowledge are present in the schematic category remain an avenue worth exploring to potentially address some spatial correlation.

Function of chariots

Lhote has argued for the war-like use of the flying gallop style based on two examples from Libya, where a flying gallop chariot is juxtaposed with a Barbary sheep hunting scene and another where a horned figure holding a javelin runs in front of the chariot (Allard-Huard, 1985: 46). However, for Muzzolini, the lone charioteer has precluded them from being interpreted as war machines, and pictorially ‘seem indeed to represent no other than the usual stereotype of the antique chariot racing’ (Muzzolini, 2001: 623). In regard to the schematic depictions of the western Saharan region, Muzzolini proposes that there is no indication of their actual function.

Lhote’s propositions about flying gallop styles progressing southward correlate with trade routes used by the Garamantes, and although Lhote does not explicitly draw on this correspondence it may serve to act as a motivation for the distribution of chariot styles. The widespread occurrence of chariot imagery on Saharan rock outcrops led to an early proposition of ‘chariot routes’ linking North and West Africa (Keech Mcintosh, 1996: 750). However, these vehicles were not suited for long-distance transport across desert terrain; more localised use is probable, operated through middlemen who were aware of the trade routes through the desert landscape.

Garamantian trade routes did not extend to the Western Sahara, Mauritania and Morocco, but there are a notable number of chariot depictions in this region (see Table 1). There is good evidence for the regional movement of copper from Mauritania to Niger between 600 and 200 BC, and the distribution of copper objects along seasonal grazing lands in Mauritania suggests that mobile pastoralists smelted and transported the copper in an exchange network (Keech McIntosh, 1996: 748). Indeed, Masonen (1997) has proposed that the Sahara has always had an interactive trade network both internally and with people from outside the continent. Rather than acting as an obstacle, the Sahara was a conduit for the movement of goods and commodities in various parts of the continent. Figure 3 shows the distribution of both schematic and flying gallop depictions in relation to known trade routes at the time.

Mattingly has observed what he terms a ‘peculiarity of the data’ (2003: 345), whereby he notes the vast majority of chariot imagery is located in the rock shelters of traditional pastoralists. These sites are away from the oases and trade routes of the ancestral Berbers, the people we may have traditionally expected to be responsible for the diffusion of chariot imagery (Mattingly et al., 2003: 345). To some degree, this may be the result of the nature of an oasis. As an area of vegetation located close to a water source, it may not provide the required rocky outcrops necessary for rock art imagery. The placement of images in rock shelters was an artistic tradition, yet may still have identified sites close to trading routes or water sources. Mapping chariot sites in relation to oases and known trade routes may go some way to helping understand any potential relationships.

The overwhelming message is that the function of chariots is unclear. Even the most well-defined and ultimately identifiable of the chariot groupings, the flying gallop, has been interpreted as either a war or racing machine. The schematic grouping is more problematic to interpret, particularly because they so noticeably lack their tractive force. By visually uncoupling, or intentionally representing the chariot as distinct from its motive power, we might surmise that the intention is to convey the importance of the technology, the display of ownership or as a sign of trade and exchange. Lhote suggests that at the time these images were made, horses were not a familiar animal in the west of the region (i.e. Mauritania, Western Sahara and the Moroccan Atlas); people used oxen instead. However, chariot images in this region are conspicuously absent of both horses and oxen. According to Strabo, the Gaetulians an ancient Berber tribe inhabiting the large desert regions south of the Atlas Mountains were known for their horse rearing, purportedly producing one hundred thousand foals a year (Lhote, 1982: 30). Although this figure may be exaggerated, it seems unlikely that horses were an unfamiliar animal. So if both horses and oxen were available, it seems an intentional artistic device to represent chariots without them.

Approaching this visual aesthetic from a Sahara-wide perspective may be missing regional or local nuances. The messages that chariots were trying to convey are likely to be multivalent and context-dependent; but in essence, potentially correlate with Muzzolini’s premise that these were differing visual strategies representing similar realities.

Anomalies

There are a few unusual representational scenes within this collection that are worth noting because they have the potential to inform our thinking on the symbolic and cognitive importance of the chariot as a cultural motif. From Algeria, two particular images stand out (see Figure 5). The first is from a site known as Ouan Mouline in the Tassili n’Ajjer in Algeria.

Figure 5.

(left) Drawing of rock art tableau showing schematic chariot adjacent to pyramidal structure. Ouan Mouline, Tassili n’Ajjer (Adapted by de Torres from Camps, 1989: 35); (right) drawing of schematic chariot with figure holding reins attached to two giraffe. (Adapted by de Torres from Camps and Gast, 1981: 77).

The scene includes a flattened two-wheeled chariot with triangular platform upon which stands a figure holding five sticks in one hand and a flail in the other. On the left of the image is a quadrangular-based structure rising to a pyramidal top, upon which sits a rooster. Two figures, one seated and one standing, are located at its base and below a figure carries a large globular pot on his head.

The structure has been compared to the Mausoleum of the Flavii at Kasserine in inland Tunisia. Built in the first half of the second century AD, the mausoleum carries an inscription that mentions a lifelike rooster that crowned the monument. Paintings from a rock cut tomb at Jbel Zabouj in northern Tunisia, datable to the third or second century BC, also depict a pointed tower topped with a cockerel, the rooster being symbolic of man’s eternal soul (Shaw, 2007: 36). In addition, the Mausoleum at Dougga in Tunisia, built in the second century BC, is engraved with a chariot just under the pyramidal top (Camps, 1989: 34). Furthermore, the Garamantes incorporated pyramid-style structures between 2 and 4 m high in their variety of funerary monuments, all of which are found within a 30 km radius of the Garamantian capital and fall within the classic Garamantian period from c.1–500 AD (Mattingly et al., 2003: 219). The association of funerary structures from Tunisia with chariot symbolism attests to the extensive communicative networks at this time and the importance of chariot imagery as a signifier of complex cultural networks.

The other image from Algeria is a schematic two-wheeled chariot adjacent to a figure holding two long sticks and four reins attached to two giraffe. The theme of a giraffe attached by a line or rope to a human figure is not unusual in Saharan rock art, one of the most famous sites being Dabous in Niger. It has been suggested that chariots may have been used for hunting giraffe (Muzzolini, 1992: 17) or that they may reflect a religious, mythical or cultural association as a rain animal (see van Hoek, 2003).

In Libya, an engraving of three schematic chariots with three crudely incised animals and a figure with both arms raised in the air (Figure 6) has been interpreted as ‘a proud owner showing off his three chariots and horses’ (Mattingly et al., 2003: 16), although the animals in question are not necessarily very equine-like.

Figure 6.

‘Proud owner shows off chariots and three horses?’ Jarma, Fazzan, Libya.

The rock art is located on the edge of an escarpment near the Garamantian capital in close proximity to Garamantian quarries and burial cairns (Mattingly et al., 2003: 314). Other adjacent rock art imagery shows a range of what have been regarded as masculine activities such as riding, hunting and brandishing weapons, and as Mattingly et al. (2003: 316) proposes ‘it is tempting to relate the symbolism of the engravings to the status of the burials with which they are spatially, and possibly temporally associated’. Smaller burials on the opposite side of the escarpment are associated with images of cows, with no gender distinctions. Mattingly suggests that tomb size may distinguish between male and female burials cairns which potentially extend to the adjacent rock art imagery (Mattingly et al., 2003: 316). Attributing gender distinctions to chariot imagery is interesting, particularly if a relationship can be made in other contexts with burial cairns.

Two further engravings from the eastern Aïr Mountains in Niger are interesting for their cultural hybridity. The first (see Figure 2 on the left) depicts a flattened out two-wheeled chariot and horses which are positioned back to back, on each side of the shaft. Interestingly, the charioteer adopts the posture of a flying gallop charioteer (although the horses do not), bending slightly forward and holding a stick; however, this figure is depicted very characteristically in the Niger Warrior style with a bi-lobular head. The second engraving depicts the chariot in the same way as the former, and while lacking a charioteer is adjacent to a characteristic Warrior style figure, commonly seen in Niger rock art. Both of these engravings are associated with giraffe imagery and may support the proposition of using chariots for hunting giraffe.

The association of chariots with Libyan Warrior figures in Niger, particularly adopting the posture of a flying gallop charioteer, picks up on a point raised earlier concerning artistic antecedents. The ways in which chariots are depicted and the association with other imagery is an interesting point of departure to explore in trying to understand how these vehicles are cognitively and visually conceptualised. Understanding stylistic variation at a regional level as it relates to other rock art imagery may help in understanding the aesthetic responses of artists when depicting chariots.

In varying degrees, such examples associate chariots with: architecture and symbols from other cultures and countries; giraffe, which may relate to hunting or have some mythical and/or symbolic significance; the Garamantian capital, possibly connected with wealth, ownership and constructional skill, but potentially also gender-related imagery; and cultural and representational hybridity. As such, these ‘anomalies’ provide some very interesting and diverse insights into the complex visual and cultural messages that chariots may convey.

Chariots as symbols

Notwithstanding the debates surrounding the dates chariots entered the Sahara, how long they were used or their function, the extent and abundance of chariot imagery indicates the vehicle had a considerable importance to a wide-ranging network of cultural groups. Furthermore, the ways in which chariots were represented, in profile or flattened, with horses or without, the number of wheels and/or shafts, with a standing or seated charioteer, with passengers, racing or static, probably testifies to the variety of uses and contexts in which these vehicles were engaged (see Table 5).

Table 5.

The quantities of chariot types across Saharan countries.

	Algeria (201)	Libya (62)	Niger (29)	Morocco (96)	Mauritania (125)	Mali (6)	Western Sahara (21)
Schematic	146	47	26	94	122	6	21
Stationary/equine	18	8	3	2	3	0	0
Flying gallop	37	7	0	0	0	0	0

Note: Total number of chariots in brackets.

Flying gallop

The flying gallop grouping is easy to define and identify, and is a localised phenomenon, confined to Libya and Algeria. Whatever date they were introduced into the Sahara, it seems they had meaningful associations with the Mediterranean world. The urban and architectural accomplishments of the Garamantes imitated proximate kingdoms and empires ‘in a kind of competition for prestige’ (Liverani, 2000: 26), and flying gallop chariots potentially conveyed messages about power, trade, wealth, technological capability and cultural capital.

An interesting observation at this stage is that if the Garamantes were the people to introduce chariots to the Sahara then one might expect the majority of flying gallop depictions to come from this core area. However, Table 5 shows that, in this sample, Algeria has more than three times the number of flying gallop depictions. However, such a reflection requires detailed mapping to identify more specifically the location of these images in relation to Garamantian territories.

While flying gallop chariot depictions seem to have been a relatively short-lived phenomenon, it was a vogue that probably embodied significant cachet and the nature of the depictions indicate their engagement in the cultural and social life of central Saharan peoples.

Schematic

In comparison, the chariots in the schematic category are more difficult to deal with because of their extensive distribution and heterogeneity. Those from Western Sahara and Mauritania are among the most schematic and abstract of all chariot depictions. In comparison to other examples in the schematic category, they appear to be reduced to the simplification of basic forms. While we must take into consideration artistic proficiency, not all people creating rock art would have the same skill sets; this should not necessarily prompt an assumption that artists were not familiar with the vehicles.

However rudimentary some of the schematic depictions appear, this graphic strategy of using basic feature extraction is, in cognitive visual terms, a powerful and efficient way of dealing with information with greater economy. In fact, our visual and memory neural networks are designed to process partially occluded or abbreviated objects as complete (Snodgrass and Kinjo, 1998). Extended preoccupation with both making images in a particular way and seeing depictions regularly at various sites may have acted as visual cues with equal efficiency as other more visually complex chariot imagery. As Morphy (2005: 53) observes, ‘Aesthetics can also be concerned with more abstract qualities, such as efficiency or aptness’.

In a similar vein, representing numerous two-wheeled chariots closely aligned to each other may not necessarily be indicative of quantities of chariots, but acts as an exaggeration of form. Neuroscientist Ramachandran (2003) has proposed that some art adheres to the Peak Shift effect, a neurological principle that was first observed in animal discrimination learning (Tinbergen, 1954), but relates to human pattern recognition and aesthetic preference. Ramachandran argues we are cognitively programmed to respond to the Peak Shift principle whereby ‘deliberate hyperbole, exaggeration, even distortion’ (Ramachandran, 2003) can provoke powerful neural responses.

Stationary and equine

The grouping of images that depicts chariots (both in profile and flattened) with horse and oxen, charioteers and passengers, attests more to the social uses of chariots and the importance of both equine and vehicular management. These are not portrayed as racing vehicles but as objects of transportation, that moved people (and possibly goods) around the landscape. However, both chariots and horses required attention and management to keep both animals and vehicles in peak condition.

Summary

There is no consensus on either the dating of chariot depictions, or their use. While the imagery is wide-ranging and varied, what Saharan populations adopt is, Muzzolini (1992: 30) argues, an unsuitable appliance for desert terrain, possibly mainly intended as an item of prestige or a ‘fad’. The introduction of chariots into the Sahara region, on the whole, represents ‘a minor technological episode’ and ‘the importance which archaeologists traditionally attribute to them in the classification seems out of proportion’ (Muzzolini, 1992: 29). Technologically, in an environment that was becoming increasingly arid, the chariot itself may not have found its ideal niche, but the widespread imagery arguably attests to the symbolic aspects of chariot depictions possibly more than their functional use. Indeed, as Muzzolini states (1992: 30), ‘symbols, in peoples’ lives, often weigh more than economic needs’. If this is the case, and that chariots acted more as symbolic cultural artefacts rather than functional objects, this necessarily affects the ways in which we approach rock art depictions. As such, their symbolic importance cannot be deemed out of proportion.

Within this pan-Saharan setting, chariots were very likely to be simultaneously symbolic, performative and functional. As a mode of transport, as a vehicle for racing, as an iconic indicator for trade and exchange and as a symbol of power and prestige, chariots in Saharan rock art acted as a significant prop in ‘a formalized drama demanding actors and spectators’ (Piggott, 1992: 47). With a current estimate of more than 1200 depictions of chariots across the Sahara, a systematic analysis that factors in the aesthetic nature of chariot representations may provide some insights into local and regional specificities, and start to elucidate how chariot imagery worked. Representational strategies are seldom random, and in this context images of chariots painted and engraved onto rocks across the Saharan landscape seem to have been part of an embodied, extended and distributed cognition.

Footnotes

Acknowledgements

I would like to thank Elizabeth Galvin, Curator of the African Rock Art Image project at the British Museum for supporting this investigation, Jorge de Torres for his help in providing maps, re-drawings and comments and John Giblin for his constructive comments. My appreciation also extends to David Mattingly for his permission to use a redrawing of a rock art image from his 2003 publication, and the constructive comments by the reviewers. The African Rock Art Image project is supported by the Arcadia Fund.

Classical authors and texts

Herodotus, Histories IV.183.

Pliny the Elder, Natural History, Book V Chapter 5.

Strabo, The Geography of Strabo, Book XVII Chapter 3.

Tacitus, Annals, III, IV and Histories, IV.

Ptolemaeus, Geographia, Book IV.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biography

Helen Anderson currently works on the African Rock Art Image Project at the British Museum. She specialises in the origin and development of ‘art’ from Middle Stone Age Africa to Upper Palaeolithic Europe, and is interested in the cognitive development of modern humans. Her work uses current research in neuroscience as way of gaining insights into the development of complex cognition in prehistory.

References

Aiken Littauer

MJH

Crouwel

Raulwing

(2002) Selected Writings on Chariots and Other Early Vehicles, Riding and Harness, Leiden: Brill.

Allard-Huard

(1985) Nile-Sahara Dialogues of the Rocks, III Innovative People: The Horse, Iron and the Camel, Le Caire: Études Scientifiques.

Camps

(1989) Les chars Sahariens. Images d'une société aristocratique. Antiquités Africaines 25: 11–40.

Camps G (ed.) (1993) Les Chars. Encyclopédie Berbère, XII. Aix-en-Provence: Édisud.

Camps

Gast

(1981) Les chars préhistoriques du Sahara: Archéologie et Techniques D'attelage, Aix-en-Provence: Université de Provence.

Coulson

Campbell

(2010) African Rock Art, New York: Harry N Abrams.

Gallinaro

(2013) Saharan rock art: Local dynamics and wider perspectives. Arts 2: 350–382.

Gauthier

(1999) Engraved chariots of Fazzan (Libya). Pictogram 11(1): 1–19.

Gauthier

(2011) Des chars et des Tifinagh: Étude aréale et correlations. Cahiers de l’AARS 15: 91–118.

10.

Graziosi

(1942) L'arte Rupestre Della Libia, Napoli: Edizioni della Mostra.

11.

Keech McIntosh

(1996) West African Savanna Kingdoms: The Sahara, Caravan Trade, and Islam. In: Fagan

(ed.) Oxford Companion to Archaeology, Oxford: Oxford University Press, pp. 748–751.

12.

Lhote

(1976) Les Gravures Rupestres de L'oued Djerat, Tassili-n-Ajjer, Alger: Société nationale d'édition et de diffusion.

13.

Lhote H (1982) Les chars rupestres sahariens. Des Syrtes au Niger, par ke pays des Garamantes et des Atlantes. Collection “Archéologie=Horizons Neufs”, Editions des Hésperides.

14.

Liverani

(2000) The Garamantes: A fresh approach. Libyan Studies 31: 17–28.

15.

Lorblanchet M (ed.) (1992) Rock Art in the Old World. New Delhi: Indira Gandhi National Centre for the Arts.

16.

Masonen

(1997) Trans-Saharan trade and the West African discovery of the Mediterranean. In: Sabour

Vikør

(eds) Ethnic Encounter and Cultural Change, London: Bergen, pp. 116–124.

17.

Mattingly D (2005) Garama: An ancient civilisation in the Central Sahara. World Archaeology 9. Available at: www.world-archaeology.com/features/garamantes-libya.htm (accessed 15 July 2016).

18.

Mattingly

Daniels

Dore

(2003) The Archaeology of Fazzān, Volume 1, Synthesis, London: Society for Libyan Studies.

19.

Mattingly

Daniels

Dore

(2007) The Archaeology of Fazzān. Volume 2, Gazetteer, Pottery and Other Finds, London: Society for Libyan Studies.

20.

Mattingly

Daniels

Dore

(2010) The Archaeology of Fazzān, Volume 3, Excavations of C. M. Daniels, London: Society for Libyan Studies.

21.

Mattingly

Sterry

(2013) The first towns in the central Sahara. Antiquity 87(336): 503–518.

22.

Mori

(1965) Tadrart Acacus: Arte Rupestre e culture del Sahara Preistorico, Torino: Giulio Einaudi.

23.

Morphy

(2005) Aesthetics across time and place: An anthropological Perspective on archaeology. In: Heyd

Clegg

(eds) Aesthetics and Rock Art, Aldershot: Ashgate, pp. 51–60.

24.

Muzzolini

(1992) The ‘Chariot Period’ of the rock art chronology in the Sahara and the Maghreb: A critical reappraisal of traditional views. In: Lorblanchet

(ed.) Rock Art in the Old World, Papers of the AURA Congress, (Darwin 1988), New Delhi: Indira Ghandi National Centre of Arts, pp. 9–31.

25.

Muzzolini

(1995) Les Images Rupestres du Sahara, Toulouse: Préhistoire du Sahara.

26.

Muzzolini

(2001) Saharan Africa. In: Whitley

(ed.) Handbook of Rock Art Research, Walnut Creek, CA: AltaMira Press, pp. 605–636.

27.

Piggott

(1992) Wagon, Chariot and Carriage, London: Thames & Hudson.

28.

Ramachandran VS (2003) The Artful Brain. Reith Lecture 3.

29.

Shaw BD (2007) Cult and Belief in Punic and Roman Africa. Princeton/Stanford Working Papers in Classics. Available at: www.princeton.edu/∼pswpc/pdfs/shaw/090705.pdf (accessed 1 July 2016).

30.

Snodgrass

Kinjo

(1998) On the generality of the perceptual closure effect. Journal of Experimental Psychology: Learning, Memory, and Cognition 24(3): 645–658.

31.

Tinbergen

(1954) Curious Naturalists, New York: Basic Books.

32.

van Hoek

(2003) The Saharan “Girafe a Lien” in rock art. Domesticated giraffe or rain animal? Comparing enigmatic giraffe petroglyphs from the Sahara and Namibia. Sahara. Prehistory and History of the Sahara 14: 49–62.

33.

Vernet

(1993) Préhistorire de la Mauritanie, Séepia: Centre Culturel de Nouakchott.

34.

Wilson AI (2005) Foggara irrigation, early state formation and Saharan trade: The Garamantes of Fazzan. In: Schriftenreihe der frontinus gesellschaft 26, internationales frontinus-symposium, wasserversorgung aus Qanaten – qanate als vorbilder im tunnelbau, 2–5 Oktober 2003. Luxemburg: Walferdange, pp.223–234.

35.

Werner L (2004) Libya’s forgotten desert kingdom. Aramco World: Arab and Islamic Cultures and Connections 55(3): 8–13.