Ananta in Caravelí? Polycephalic snakes in Desert Andes rock art

Maarten van Hoek



Sometimes it is surprising to see how things seem to be related. Namely, in order to explain the subject of this paper about certain rock art images in the Andes, South America, I first take the interested reader to the Asian subcontinent of India where ‘Ananta’ is a most important religious Hindu concept. ‘Ananta is a Sanskrit term, which means ‘endless’ or ‘limitless’. It also means ‘eternal’ or ‘infinite’. ‘Ananta’ is also known as the ‘Shesha-naga’, the celestial snake and as such ‘Ananta’ is also celebrated as the serpent of infinity.  Shesha’ is sometimes shown as a five-headed or seven-headed, but more commonly as a many thousand-headed serpent (Wikipedia).


This sounds interesting, but what is the relationship between the ‘thousand-headed serpent’ of the ancient Indian pantheon and the Andes of South America; an area 17.000 km west of India and separated by vast oceans?


The purely coincidental ‘relationship’ is that the geographical site of Ananta exists in the deserts west of the Andes. Ananta is the name of a Hacienda but also of a nearby prehistoric petroglyph site in the valley of the river Caravelí in the Department of Arequipa, southern Peru (there are more sites [including a mountain and a river] that are called Ananta in southern Peru). The more perplexing but still fortuitous ‘relationship’ is that the river valley in which Ananta is situated, the valley of the Río Caravelí, proved to be lined with rock art sites featuring remarkable and exceptional snake imagery, including a relatively large number of most mystifying images of polycephalic snakes. The array of petroglyphs of (especially polycephalic) snakes in the Caravelí Valley proved to be so atypical that it is necessary to first put the imagery of (polycephalic) snakes in a proper Andean context.



Snake Imagery of the Desert Andes


Snake symbolism and therefore also snakes imagery are rather common and important in many ancient Andean cultures. Therefore it will be no surprise that also the rock art of the Desert Andes (the extremely dry coastal strip west of the High Andes from Olmos in the north of Peru to San Pedro de Atacama in northern Chile) often features snake imagery. Remarkably however, there is a larger quantity of snake petroglyphs in the Ica-Arequipa area of southern Peru. To the north of this mountainous area, snake imagery occurs less frequently in rock art and, more surprising, when going south, entering the Atacama Desert of northern Chile, the occurrence of snake imagery in rock art seems to peter out rapidly.


Surprisingly however, the rock art of the Department of Arequipa has an above average occurrence of snake imagery, mainly in the coastal desert between the valleys of the river Atico and the river Tambo. This region is the Study Area of this paper; an area of about 200 km by 100 km, encompassing the drainages of the Caravelí River (the Caravelí Valley is only a small part of the much larger Province of Caravelí) and of the Manga, Majes, Sihuas and Vitór rivers (Figure 1). The major concentration of snake petroglyphs occurs in the Majes Valley. But there are two valleys that have more amazing snake content. Those two valleys - Vitór in the east and, in particular, Caravelí in the west - also form the limits of the Study Area. In order to protect vulnerable and still unreported rock art sites, I only will mention the valleys, not the sites or the locations, in this paper. The sites in Caravelí will for the same reason be numbered: Caravelí 1, 2 and so on, without revealing their names and locations (please, contact the author for further information).


Figure 1. Map of a large part of the Department of Arequipa, southern Peru, with the location
of the relevant river valleys of the Study Area (all rock art sites have been omitted).
Drawing © by Maarten van Hoek,
based on Google Earth Relief Maps



Polycephalic Snake Imagery in Andean Rock Art


Although multi-headed creatures, especially snakes, are abundantly known in many mythologies across the world, most readers will consider the existence of a thousand-headed serpent as pure fantasy. However, in real life polycephalic animals do occur (Wikipedia). Polycephaly is the biological condition of one creature having more than one head. Yet, in the real world only bicephalic or tricephalic animals are the only types of multi-headed creatures that can occur, which, in the case of snakes, means that they may sometimes have two heads and, even more rarely, three heads. But a real snake with more than three heads seems to be an absurdity (internet). Importantly however, polycephalic snakes in the real world only may have two (or three) heads next to each other, thus emerging from the one and only neck the creature has. Never has a real snake been reported that has a second head at its tail-end (disregarding the fact that in that case there is no tail-end).


Although headless (unfinished?) snake images sometimes occur, especially monocephalic examples are the most common snake representations in the rock art of the Desert Andes. Understandably their appearance varies enormously. The simplest shape is the single zigzagging or serpentine groove with often a somewhat larger depression representing the head. More complex outlined examples often have interior decoration (dots, circles, [zigzag] lines and so on), while the head may show facial features and have several appendages and/or a tongue sticking out. A special group of snake images are the so called ‘feathered’ snakes (for an example see Van Hoek 2014: Figs 7 and 13).


Also the size varies enormously. In Majes snake petroglyphs have been reported ranging from a few decimetres to a three meters long bicephalic snake. Even a monstrous snake of 15 meters in length has once been reported in Majes, but unfortunately that alleged petroglyph has never been described in a scientific publication.


More interesting in the scope of the current paper are the snake images that have a snake head at one end and an unexpected feature at the other (tail) end. For instance, a snake petroglyph in Majes (Figure 2) has a solar symbol at its tail-end. A monocephalic ‘snake’ with a human-like heads occurs at a site in the Study Area, while outside the Study Area there is at least one rock painting of a bicephalic snake with a human-like head at one end. It now proves that the most common type of abnormal snake imagery concerns bicephalic snakes; a snake with a second snake head at the purported tail-end.


Figure 2. Petroglyph of a snake in Majes. Photograph © by Maarten van Hoek.



Bicephalic Snakes

Bicephalic snakes in rock art are known to me to have been reported from the north of Peru to - across the Andes - the west of Argentina. They also appear on textiles and ceramics. Although their scarce manifestations may differ quite a lot, all examples outside the Study Area - known to me - invariably only have a second snake head at the tail end. The second head may often lack facial features; then only the shape of a head is drawn.


In the Chancay-Reque Valley of northern Peru a fine U-shaped bicephalic snake petroglyph has been reported (Van Hoek 2012a: 126), while another U-shaped bicephalic snake petroglyph from the Jequetepeque Valley, also in northern Peru, seems to be involved in a fertility scene (Van Hoek 2012b: Fig. 156). In the Palpa and Las Trancas valleys in the Ica-Nazca area of southern Peru a few bicephalic snake petroglyphs seem to have small bird-wing-like appendages (Figure 3), while sometimes they also have human-like faces (Figure 4). South of the Study Area bicephalic snake imagery is (very) rare in the Desert Andes. I know of one example in the Huatacondo Valley in the north of Chile, while in the Loa Valley, also in northern Chile, two bicephalic snakes (or perhaps four monocephalic snakes) seem to emerge from the belt of a possible deity petroglyph (Chiuchiu - Van Hoek 2001: Fig. 6).

Figure 3. Petroglyph of a snake in the Palpa Valley, Departamento de Ica, Perú.
Drawing © by Maarten van Hoek. Any inaccuracy is my responsibility.

Figure 4. Petroglyph of a snake in the Las Trancas Valley, Departamento de Ica, Perú.
Drawing © by Maarten van Hoek, based on a photograph posted by the Municipalidad
Valle Las Trancas in Flickr 2009)
. However, any inaccuracy is my responsibility


In contrast, relatively many examples of bicephalic snake petroglyphs occur in the Study Area. Except for some notable exceptions that are the subject of this paper, all have the second head only at the tail-end. Especially in Majes large and/or weird-shaped bicephalic snake petroglyphs have been reported, such as a completely straight and still undulating example and another example of more than three metres (possibly symbolically beheaded).


Almost every example of the (polycephalic) snake imagery in the Study Area displays the heads ‘frontally’, which means that the two eyes (and sometimes other facial features) and the centrally positioned tongue are observed as if the head is seen from above; the so called bird’s eye view. There are unparalleled exceptions however. A large bicephalic snake petroglyph, also in Majes, shows a very large, horned head depicted in profile and a very small second head at the tail end that is also shown in profile. A specific class of profile-(double)-headed creatures will be discussed in another paragraph.


As it is not the intention of the current paper to discuss all (bicephalic) snake images of the Study Area, I now focus on the more exceptional images.


There are two valleys in the Study Area that have an exceptionally high number of polycephalic snake petroglyphs. The eastern limit of the Study Area is the Vitór drainage. In a relatively small part of this drainage (called the Río Chili at that point) is a major concentration of petroglyph sites that are characterised by very specific snake imagery. For matters of convenience I regard all snake-like biomorphic petroglyphs in the Study Area to represent or symbolise real snakes, acknowledging, however, that other creatures may have been intended by the ancient manufacturers. The Vitór snake petroglyphs are not always zigzagging or meandering but often have an irregular but angular appearance (Figure 5). Also the heads have been drawn in an often angular way. Many of those specific snake petroglyphs are bicephalic and are moreover characterised by rows of often short, triangular appendages. For this reason images of such creatures have also been interpreted as centipedes. This type of snake has certainly been recorded once in Majes and - possibly - one time in Ocoña.


Figure 5. Petroglyph of a snake in Vitór. Drawing © by Maarten van Hoek, based on a photograph by Augusto Cordona Rosas (In You Tube - 2008). However, any inaccuracy is my responsibility.



There is at least one petroglyph in this Vitór group that shows an abnormality. The second head of this specific snake petroglyph does not appear at the other (tail) end, but emerges from an apparently randomly chosen spot of the larger snake’s body (Figure 6). In fact two snakes may have intentionally been merged into one complex, bicephalic creature, which, from a distance, vaguely resembles an outlined cross; a Chacana perhaps (the third, parallel groove in Figure 6 may have been added later). It is certain that this part of Vitór represents a very special place. However, in the Caravelí Valley, 180 km NW of Vitór, even more intricate, polycephalic snake petroglyphs once were manufactured.


Figure 6. Petroglyph of a snake in Vitór. Drawing © by Maarten van Hoek, based on a photograph
kindly supplied by the
Oficina de Imagen Institucional de la Municipalidad de La Joya,
Departamento de Arequipa, Perú. However, any inaccuracy is my responsibility


Caravelí Rock Art

The oldest record of rock art in the Caravelí Valley - although not mentioned or cited in any of the publications that I have available or consulted - dates from 1911 (also mentioning the representation of a snake). In 1911 the well known American explorer and discoverer of Machu Picchu, Hiram Bingham, includes the following record about his journey from Caravelí to Ocoña (1912; 1922):

Leaving the pleasant shade trees of Caravelí, we climbed the barren, desolate hills of coarse gravel and lava rock and left the canyon. We were surprised to find near the top of the rise the scattered foundations of fifty little circular or oval huts averaging eight feet in diameter. There was no water near here. Hardly a green thing of any sort was to be seen in the vicinity, yet here had once been a village. It seemed to belong to the same period as that found on the southern slopes of the Parinacochas Basin. The road was one of the worst we encountered anywhere, being at times merely a rough, rocky trail over and among huge piles of lava blocks. Several of the larger boulders were covered with pictographs. They represented a serpent and a sun, besides men and animals.

Unfortunately, it is still unknown where this site is actually located. After the discovery by Hiram Bingham the Caravelí Valley became a neglected region concerning rock art research. In a publication by Linares Málaga only two (different) sites were marked on his map of rock art sites in Arequipa (1973: Figura 58; No. 16 and 39). The more recent Inventario Nacional published by Rainer Hostnig (2003) altogether mentions six sites in Caravelí, mainly briefly referring to older publications by Linares Málaga and not mentioning Hiram Bingham’s discovery. The Resolución Directoral Nacional No. 380, published by the INC - Instituto Nacional de Cultura de Perú - (INC 2010) declared that from 2010 five rock art sites in the Caravelí District should be considered to be Cultural Legacy of the Nation (Patrimonio Cultural de la Nácion). Regrettably, this INC publication only reveals the names and the UTM locations of five sites but does not provide any further details.


Because of the possible double-entries and the several alternative local names for the possibly up to 19 (!?) rock art sites in the Caravelí Valley a most confusing situation emerged. Moreover, the information that came with the photographs also might include double-entries or mixed-up names. Because I alone am responsible for the information presented in this paper, I will not use any site names. Alternatively, I will assign a number (for instance Caravelí 1) to the purported eight rock art sites in the Caravelí Valley of which I have a modest photographic record. The other sites, recorded by Hiram Bingham (1911), or reported by the INC (2010) or in Rainer Hostnig’s inventory (2003), will not be considered here due to lack of information.


As only very little has actually been published about Caravelí rock art, I only know the (often much) approximated locations of most sites. Google Earth shows very many areas in the valley with boulders and outcrops that could offer suitable surfaces for rock art production. Also, Google Earth shows an enormous amount of prehistoric remains both very near the river as well as at high altitudes on the surrounding High Pampas. Yet, based on the available information, I could pinpoint the exact locations of only a limited number of sites/panels.

However, I must emphasise here that I never visited the Caravelí Valley myself. All my observations in the current paper are based on the personal photographic records generously made available to me by
Rainer Hostnig, Leonel Edson Guerrero Rumaldo, Mario Antonio Casas Berdejo, Katherine Solórzano (2008 to 2014: pers. comm.) and on other sources that I have at hand (mentioned in this paper). However, only I am responsible for all my illustrations, observations, conclusions and theories presented in this paper.

As I definitely have no complete photographic record of any of the (purported) 19 rock art sites in the Caravelí Valley, the only goal of this paper is to offer an impression of the enormous and unexpected diversity of snake imagery in this area (often without being able to offer a scale as my surveys only involve photography). In other words, it is not at all my intention to offer a complete scientific inventory of any rock art site in Caravelí.


Monocephalic snakes

Monocephalic snake imagery in the Caravelí Valley ranges from very simple, single groove examples to highly complex, delicately outlined configurations. At Caravelí 1 several rather deep grooves, some quite straight, others meandering, seem to represent snakes. Some have a cupule at one end that may represent the head. Similar petroglyphs have been reported at Caravelí 2 (Figure 7), Caravelí 4 and Caravelí 7.

Figure 7. Petroglyph of a snake at Caravelí 2.
Photograph © by Leonel Edson Geurrero Rumaldo.


Outlined examples (some possibly headless) also occur at Caravelí 1 and 8, while at Caravelí 3 a very long snake petroglyph is curled up on a boulder (Figure 8). The strangely shaped head seems to have its tongue sticking out. At the same site is a panel with a long snake petroglyph, its meandering body bent in a distinct 90˚ turn. Also this petroglyph has a large, strange head with human-like facial features (Figure 9). A very long, meandering snake at Caravelí 5 has a ‘horned’ head (Figure 10). At Caravelí 6 an irregular, outlined snake-like petroglyph seems to have solar head with nine short rays emerging from the circular head that encloses two or three dots, which may represent the eyes/nose. A boulder at Caravelí 7 has three rather straight, still outlined petroglyphs that look like snakes composed of ‘strings of beads’. One such pattern seems to have a small triangular head with a short tongue sticking out (blue arrow in Figure 11). A boulder at the same site has a petroglyph of a snake that has the shape of the Andean double-stepped pattern.


Figure 8. Petroglyph of a large snake at Caravelí 3.
Photograph © by Leonel Edson Geurrero Rumaldo; digitally enhanced by the author

Figure 9. Petroglyph of a snake at Caravelí 3.
Figure 10.
Detail of the head of a petroglyph of a long snake at Caravelí 5. Both drawings
© by Maarten van Hoek, based on photographs by Leonel Edson Geurrero Rumaldo
However, any inaccuracy is my responsibility

Figure 11. Petroglyphs of possible snakes at Caravelí 7.
Photograph © by Mario Antonio Casas Berdejo.


More complex and mystifying snake petroglyphs include the slightly meandering body of a snake (?) at Caravelí 2. From both sides of the ‘body’ several rather deep, more or less parallel grooves emerge, while two large circles with a central dot each attached to one side the body look like two menacing eyes (in an otherwise illogical position).

Also at Caravelí 2 a (large?) petroglyph seems - at first sight - to represent an anthropomorph, its head crowned by a floating semi-circular device of unknown meaning. However, it has a clearly meandering and relatively long body that may well symbolise a snake (Figure 12). Also, the simple outlined head with three cupules representing the eyes and mouth may be compared with similar heads on more natural looking snake images. Moreover, the serpentine body is decorated with a row of eight circles with a central dot and - at the upper end - a cupule; a feature often seen in snake petroglyphs in the Study Area. The legs and arms may have been added in order to symbolically express the bond between humans and snakes. At the lower end of the meandering body is a feature that is hard to determine. It may represent a human sexual organ, but, if so, it cannot be decided whether it has a male or female content.


Figure 12. Petroglyph of a snake-anthropomorph at Caravelí 2. Drawing © by Maarten van Hoek, based
on a photograph by Leonel Edson Geurrero Rumaldo
. However, any inaccuracy is my responsibility



A completely different but also unique petroglyph of a snake occurs at Caravelí 7 (Figure 13). An enormous head (measuring about 70 cm across the eyes) is shaped like the well-known Andean double-stepped pattern. It has two circles with central dot that clearly represent the eyes, which are connected by a straight, single groove to an outlined area what may represent a mouth. From this outlined area a semi-circle with three short rays emerges. From the head four converging, double-outlined grooves of about 100 cm in length (the complete image has an overall length of just over 180 cm) form the wavy body of this grotesque snake petroglyph (subjective term).


Figure 13. Petroglyph of a snake at Caravelí 7. Drawing © by Maarten van Hoek, based on
a photograph by
Mario Antonio Casas Berdejo. However, any inaccuracy is my responsibility



The Bicephalic Snake Petroglyphs of the Caravelí Valley

The more standard class of polycephalic snake imagery concerns the snake with two heads; one at either end of the body. An enormous boulder at Caravelí 7, decorated on several sides, features a number of images of monocephalic snakes of much different shapes and lengths (and many other images). On the upper panel is a large U-shape bicephalic snake (about 60 cm), both heads with tongues sticking out (Figure 14). The much weathered body seems to be decorated with segments and/or dots. At the same site is another large boulder with some most interesting petroglyphs. One image depicts a bicephalic snake shaped like the letter S (Figure 15). The body is filled with a row of dots. Each head has two backwards-curving ‘horns’ (single grooves). More surprisingly, each head has an extra extension in front of the head: a ladder design and a grid design; possibly representing the tongues.


Figure 14. Petroglyph of a snake at Caravelí 7. Drawing © by Maarten van Hoek, based on a
photograph by
Mario Antonio Casas Berdejo. However, any inaccuracy is my responsibility

Figure 15. Petroglyph of a snake at Caravelí 7. Drawing © by Maarten van Hoek, based on a
photograph by
Mario Antonio Casas Berdejo. However, any inaccuracy is my responsibility



Also at Caravelí 7 is a boulder that has a large petroglyph that may be regarded as a bicephalic snake. It is a linear creature (about 64 cm in length) completely drawn in outline that has a (feline?) head at each end, but this time both heads have been depicted in profile (Figure 16A). The upper and lower lips are curled, as are the (feline?) ears. The open mouths show rectangular teeth. The body is decorated with some curls and grooves. This petroglyph clearly is of a different class and is paralleled at only two other sites. A bicephalic example from Ocoña (Figure 16B) was reported to me by Rainer Hostnig (2008: pers. comm.; Van Hoek 2011: Fig. 70), while a monocephalic creature from Majes (Figure 16C) was reported as early as 1956 (Linares Málaga 1990: Van Hoek 2011: Fig. 72; 2013: Fig: 73).


Figure 16. A: Petroglyph of a snake (?) at Caravelí 7. Drawing © by Maarten van Hoek, based
on a photograph by
Mario Antonio Casas Berdejo. B: Petroglyph of a snake (?) in Ocoña.
Drawing © by Maarten van Hoek, based on a photograph by
Rainer Hostnig.
C: Petroglyph of a snake (?) in Majes. Drawing © by Maarten van Hoek.
Any inaccuracy is my responsibility



The Polycephalic Snake Petroglyphs of the Caravelí Valley

The most unexpected outcome of the study of the photographic records of several rock art sites in the Caravelí Valley was that there are petroglyphs of snakes that have more than two heads. This seems impossible, but, apart from two tricephalic snake petroglyphs (that just possibly may once have been seen in real live in the Caravelí Valley), it was - for some unknown reason - crucial for the ancient peoples of this valley to manufacture weird-looking (subjective term) configurations of snakes with four heads, with six heads and even with seven heads. Even though, so far, only one or two examples of each instance have been recorded, the concentration of such exceptional snake representations in only one valley must have a special meaning that, unfortunately, is still unknown. But something special must have triggered the manufacture of bizarre (subjective term) snake imagery in this valley that - as far as know - is unparalleled in Andean iconography.



Tricephalic Snake Petroglyphs

Caravelí 7 has two large boulders, each with one petroglyph of a tricephalic snake (and many other petroglyphs, including snakes). A vertical panel of one colossal boulder is completely covered with an enormous amount of (often superimposed) petroglyphs. More or less in the centre of the panel is a rather deeply carved configuration of three upwards looking snakeheads, each with a very short neck that are connected below the central head (Figure 17). Below the point of convergence is a circular pattern that seems to be part of the whole configuration. Each head has two cupules that represent the eyes and an extra cupule possibly representing the mouth. From the mouth area of each head two short, curved lines, delicately executed, emerge. In at least one case those two grooves seem to have been incised. Possibly the others were initially incised as well, but they seem to have been pecked over. Although the configuration may be considered as the merging of two bicephalic snakes (both U-shaped), it is unquestionable that the arrangement of this trinity is premeditated.


Figure 17. Petroglyph of a snake at Caravelí 7. Drawing © by Maarten van Hoek, based on a
photograph by
Mario Antonio Casas Berdejo. However, any inaccuracy is my responsibility



Intention also seems to be the case with the other example at Caravelí 7. On the large, upwards facing panel of another large boulder is a confusion of many small quadrupeds (probably camelids), lines, symbols (for instance a solar symbol) and some anthropomorphs. Although there are a few simple, single line serpentine grooves (probably snakes) there also is an outlined, tricephalic snake petroglyph that is hard to discern in the confusion of lines (Figure 18). I consider the horizontally arranged snake as the main body. It is a more or less straight body that ends in a triangular head enclosing a circle and a small arc (the mouth?). The main body is crossed (cut?) by a straight groove that may well be a later addition. From the main body two similar snakes emerge, each ending in a roughly triangular head. The upward pointing snake has similar facial features and two circles on the body. The body of the downward pointing snake is empty, while its head is superimposed by a groove that seems to represent a corral around a herd of camelids.


Figure 18. Petroglyph of a snake at Caravelí 7. Drawing © by Maarten van Hoek, based
on a photograph by
Mario Antonio Casas Berdejo. However, any inaccuracy is my responsibility



The Four-Headed Snake

Also at Caravelí 7 is a boulder with a remarkable configuration of two bicephalic snakes that are unmistakably and intentionally connected, possibly in order to create a four-headed snake configuration (Figure 19). It is even possible to consider the cross-shaped connecting pattern in the centre of the configuration to represent the merged heads of two snakes. Then a six-headed creature emerges. The whole configuration is clearly angular (compare this with a snake on a Siguas textile illustrated by Joerg Haeberli 2001: Fig. 7). Also the heads are angular and only two heads have simple dots that represent the eyes. Again the petroglyph is large. It measures about two metres from left to right.


Figure 19. Petroglyph of a snake at Caravelí 7. Drawing © by Maarten van Hoek, based on a
photograph by
Mario Antonio Casas Berdejo. However, any inaccuracy is my responsibility




The Six-Headed Snake

At Caravelí 8 is a boulder with a few petroglyphs including one remarkable snake petroglyph (Figure 20).  The outlined main body is more or less U-shaped and encloses some dots while the ends are open. Separated from the main body is the principal head with two dots representing the eyes. The mouth area is open. The second head at the tail-end is also separated from the main body and has a natural depression in its centre and it is also open-ended. From the main body (but still separated from it) the short bodies of four snakes emerge, all with dots for eyes. Except for one instance, all the heads of this six-headed snake are separated from the bodies and only one has a closed semi-circle for a head.


Figure 20. Petroglyph of a snake at Caravelí 7. Drawing © by Maarten van Hoek, based on a
photograph by
Mario Antonio Casas Berdejo. However, any inaccuracy is my responsibility



The Seven-Headed Snake


It seems impossible, but a boulder at Caravelí 4 features a seven-headed snake petroglyph (Figure 21). The main body consists of two parts. One half of the body is slightly meandering and rather wide, while the tail is narrower, straight and is positioned at a 90˚ angle. The main body ends in a relatively large circular head with two dots for eyes (the mouth area seems to be damaged and therefore may have incorrectly been depicted in Figure 21). From the principal head run two short grooves that both ends in a simple curl. The tail-end has a much smaller circular head with possibly three dots that represent the facial features. From its head two short slightly curved grooves run upwards. To the left of the tail are two similarly narrow and straight snakes that are attached to the main body. One of those two snakes has a distinct circular head with two dots for eyes and a short groove representing the mouth, thus creating a human-like face. From the head two short curved grooves emerge. The other snake may have had a head, but that area of the rock’s surface is rough and this snake may never have had a head.


Figure 21. Petroglyph of a snake at Caravelí 4. Drawing © by Maarten van Hoek, based on a photograph
by Leonel Edson Geurrero Rumaldo
. However, any inaccuracy is my responsibility



At the other side of the main body three similarly narrow and straight snakes are attached to the main body pointing in the opposite direction. The central example of those three snakes has a circular head with two dots for eyes. From the head three, possibly more, short grooves emerge. Probably because of lack of space on the panel the shortest of the three snakes is not that well elaborated. It has an indistinct circular head with possible facial features. The last snake also seems to have been unfinished as in the rough area of the rock’s surface no snake head can be discerned anymore. Although no factual seven heads are visible at the moment, it is quite certain that a seven-headed creature has been intended.




Although statistically the Caravelí Valley definitely has not the biggest number of snake petroglyphs in the Study Area, it is undeniable that the valley stands out regarding the graphical content of and diversity in snake imagery. Some of the Caravelí snake images are even unparalleled in Andean rock art. Although this paper is based on an incomplete photographic record, the still relatively high number of images of polycephalic snakes clearly demonstrates that the Caravelí Valley represents a special place in Andean worldview. It is however a mystery why such specific images have so far only been reported from one single valley in the Desert Andes. Further research may reveal similar imagery in Caravelí and perhaps elsewhere in the Study Area (and beyond). Additional contextual information from the many archaeological remains in the Caravelí Valley may shed light on the mystery and perhaps it will one time be possible to explain the ‘Ananta’ phenomenon of the Caravelí Valley.





This paper could never have been written without the invaluable help and assistance of the following people: Rainer Hostnig from Cusco, Peru; Leonel Edson Guerrero Rumaldo from Arequipa, Peru; Mario Antonio Casas Berdejo from Arequipa, Peru; and Katherine Solórzano of the Oficina de Imagen Institucional de la Municipalidad de La Joya, Arequipa, Peru. All those people generously made available to me their (personal) photographs of Arequipa rock art and  granted me permission to use their graphic material. Moreover, they often offered me useful additional information. However, I would like to emphasise again that only I am responsible for all my illustrations, information, observations, conclusions and theories presented in this paper (and, of course, in my earlier publications).


¿Preguntas, comentarios? escriba a:

Cómo citar este artículo

van Hoek, Maarten. Ananta in Caravelí? Polycephalic Snakes in Desert Andes Rock Art.
En Rupestreweb,



(The underlined blue references are hyperlinks as accessible in 2014)


Bingham. H. 1912. The Ascent of Coropuna. Harper's New Monthly Magazine. Vol. 125; Number 741-45: pp 489–502. Henry Mills Alden, editor.


Bingham. H. 1922. Inca Land, Explorations in the Highlands of Peru. The Riverside Press Cambridge. Boston and New York.

Haeberli, J. 2001. Tiempo y Tradición en Arequipa, Perú, y el Surgimiento de la Chronología del Tema de la Deidad Central. Boletín de Arqueología. PUCP, No. 5, pp 89 - 137.

Hostnig, R. 2003. Arte rupestre del Perú. Inventario Nacional. CONCYTEC, Lima, Perú.


I.N.C. 2010. Resolución Directoral Nacional No. 380 / INC. Lima, Peru.

Linares Málaga, E.1973. Anotaciones sobre las cuatro modalidades de arte rupestre en Arequipa (pictografías, petroglifos, arte rupestre mobiliar y geoglifos). In: Anales Cientificos de la Universidad del Centro del Perú. Corrected version in: Boletín APAR. 2013. Vol. 4, No 15-16.

Linares Málaga, E.  1990. Prehistoria de Arequipa. Vol. 1. CONCYTEC-UNSA, Arequipa, Perú.

Van Hoek, M. 2001. The hidden gods of Chiuchiu. Boletin-e AZETA. Museo Arqueológico San Miguel de Azapa, Arica, Chile.

Van Hoek, M. 2011. The Chavín Controversy - Rock Art from the Andean Formative Period. Oisterwijk, The Netherlands. Privately published using the BLURB Creative Publishing Service.

Van Hoek, M. 2012a. Cerro Mulato: Rock Art of the Reque-Chancay Drainage, Peru. Oisterwijk, The Netherlands. Privately published using the BLURB Creative Publishing Service.

Van Hoek, M. 2012b. Rumimantam Llaqllasaq Wirpuykita: The ‘Cycle of Life’ in the Rock Art of the Desert Andes. Oisterwijk, The Netherlands. Privately published using the BLURB Creative Publishing Service.

Van Hoek, M. 2013. The Carcancha and the Apu. Rock Art in the Death Valley of the Andes. Oisterwijk, The Netherlands. Privately published using the BLURB Creative Publishing Service.

Van Hoek, M. 2014. A Critical Analysis of the Rock Art on Boulder CNG-020, Cerro Negro, Chicama, La Libertad, Peru. In Rupestreweb.