Figure 1. Equine engraving on rib fragment, presumed to be from the Creswellian of Robin Hood’s Cave, Derbyshire, England, excavated in 1876.


Figure 2. Drawing of the same object.


Figure 3. The first version of the Church Hole main figure (a), published in 2003, and the second version of the same figure (b), published in 2004, by Ripoll et al.


Figure 4. The third version of the same figure, published by Ripoll et al. in 2005.

The following report has appeared in Spanish on a Colombian rock art site, Rupestreweb. The English version is an attempt by R. G. Bednarik to translate it (but incorporates welcome improvements by Toni Evans).


On 14 April 2003, an interdisciplinary team directed by Dr Sergio Ripoll, Director of the Laboratory of Palaeolithic Studies (Laboratorio de Estudios Paleolíticos), attached to the Research Institute of the Universidad Nacional de Educación a Distancia (Spain), in collaboration with the noted English researchers, Dr Paul G. Bahn, freelance, and Dr Paul Pettitt of the University of Oxford, carried out the first discovery of Palaeolithic rock art in Great Britain.

Palaeolithic rock art occurs widely in the Old Continent, although for a variety of reasons it is absent in particular areas. The main one is the existence of a thick layer of ice of variable thickness that covered the north of Europe from Dublin to Moscow, extending at times to the present cities of Manchester, Leipzig and Warsaw. In other cases it is due to the lack of caves, or the geological conditions of the land. Lastly, there are other areas, such as the British Isles where, in spite of an abundance of cavities, so far no finds of this type were known in the area that was not occupied by the ice.

Due to this inexplicable hiatus we organised a visit to try to check, in some cavities of various areas, if the geological or environmental conditions or man’s own agency had been the causes of the lack of such rock art. With this objective we examined the groups of Creswell Crags (Derbyshire), Paviland (Wales), Kent’s Cavern (Cornwall) and Gough’s Cave (Cheddar). In all of them the existence of archaeological deposits of the Upper Palaeolithic had been documented. For the time being the only place where there is clear evidence of rock art is in some of the caves of the group of Creswell Crags. On the other hand the only pieces of portable art known in the United Kingdom also come from these caves: the horse from Robin Hood Cave and the anthropomorph of Pin Hole Cave. They are both attributed to the Creswellian, a local culture comparable with the Upper Magdalenian dated to about 12 000 years ago.

Setting out with the idea that it was hardly probable that any type of painted representations would exist, we concentrated on the search for engravings, which are generally more difficult to observe and identify. Nevertheless we were well aware that all these cavities had already been studied by noted investigators for almost one century and half.

With the appropriate system of illumination and the experience accumulated by the study of numerous groups of Palaeolithic art in the Iberian Peninsula, we centred our attention on the group of cavities of Creswell Crags in Derbyshire. Immediately we identified numerous engraved lines, very damaged by fairly modern graffiti. These were mainly in Robin Hood Cave and in the nearby cave of Mother Grundy’s Parlour, where we have located several incised lines of difficult interpretation and, inside a short gallery, the fore-section of a horse engraving in quite a shallow line. These groups are being studied in depth at the present.

The most spectacular discovery is the one carried out in the cave of Church Hole, on the south side of the group of cavities of Creswell Crags. This cavity was dug in the 1870s by the Rev. Magens Mello and Sir William Boyd Dawkins. The recovered materials were classified by D. Garrod in 1926 as Creswellian and they are now related to numerous radiocarbon dates that locate this cultural horizon between 12 500 and 12 000 B.P.

In this cave we have now identified two decorated panels. In the first of them is recognised a male goat toward the left. The representation is 57.2 cm long from the muzzle to the croup, and 40.4 cm wide from the extremity of the horn to the end of the foreleg. It appears in semi-twisted perspective, that is to say the two horns are depicted, but of the rest of the duplicated elements, that is to say legs, ears or eyes, only one of them is shown. The groove, totally smooth, is relatively wide and not very deep.

In the second panel, located some 25 metres from the mouth, we have found two other engraved figures that possibly represent two birds facing each other. The one on the right only presents the area of the head and it possesses a very long neck, which one could associate with a crane-like bird or a swan. The silhouette on the left, with very elongated sub-triangular tail, and a head that seems to present a curved beak, we could interpret as a bird of prey. The dimensions of these representations are around 30 cm wide by 32 cm long.

Stylistically all these figures can be placed in Style IV of André Leroi-Gourhan and as far as they can be dated without an exhaustive study, they are of the final phase of the glacial period, that is to say between 15 000 and 10 000 years before the present.

Although an initial study, and in the hope of new findings appearing soon, our discovery finally places Great Britain on the map of distribution of Palaeolithic rock art. Up to now the most northern example was the cave of Gouy, near the estuary of the river Seine in France, but Church Hole is approximately 500 km more to the north and this confirms the great importance and potentiality of the one group of Creswell Crags. This discovery opens up, also, new roads of study on the ways of life of some people that inhabited the very area next to the perpetual glaciers that covered the north of Europe. If the Cueva del Moro at Tarifa (Cadiz) that we found in the year 1994 is assumed to be the southernmost Palaeolithic art of Europe, now in 2003 Church Hole constitutes the northernmost Palaeolithic art of the Old Continent.

Sergio Ripoll, Paul Bahn and Paul Pettitt



El 14 de abril de 2003 un equipo interdisciplinary dirigido por el Dr. Sergio Ripoll director del Laboratorio de Estudios Paleolíticos (L.E.P.) integrado en el Instituto de Investigación de la Universidad Nacional de Educación a Distancia, en colaboración con los prestigiosos investigadores ingleses Dr. Paul G. Bahn, freelance, y Dr. Paul Pettitt de la Universidad de Oxford, realizó el primer descubrimiento de arte rupestre paleolítico en Gran Bretaña.

El arte rupestre paleolítico está prácticamente generalizado en todo el Viejo Continente, aunque en determinadas zonas por varias circunstancias está ausente. La principal se debe a la existencia de una espesa capa de hielo de espesor variable que cubrió el Norte de Europa desde Dublín hasta Moscú llegando en algunos momentos hasta las actuales ciudades de Manchester, Leipzig o Varsovia. En otros casos se debe a la inexistencia de cuevas, por las condiciones geológicas del terreno. Por último hay otras zonas como las Islas Británicas donde, a pesar de existir abundantes cavidades, no se conocían hasta el momento representaciones de este tipo en el área que no estuvo ocupada por el hielo.

Debido a lo inexplicable de este vacío organizamos una visita para intentar comprobar, en algunas cavidades de diversas zonas, si las condiciones geológicas, medioambientales o el propio hombre habían sido las causas de su inexistencia. Con este objetivo prospectamos los conjuntos de Creswell Crags (Derbyshire), Paviland (Gales), Kent’s Cavern (Cornualles) y Gough’s Cave (Cheddar). En todas ellas se había documentado la existencia de depósitos arqueológicos encuadrables en el Paleolítico Superior. Por ahora el único lugar en que hay una clara evidencia de arte rupestre, es en algunas de las cuevas del conjunto de Creswell Crags. Por otra parte las únicas piezas de arte mueble conocidas en el Reino Unido proceden de estas cuevas: el protomos de caballo de Robin Hood Cave y el antropomorfo de Pin Hole Cave. Ambos se atribuyen al Creswelliense, cultura local paralelizable con el Magdaleniense Superior fechada en hace unos 12.000 años.

Partiendo de la idea de que parecía poco probable que existiera algún tipo de representación pintada, nos centramos en la búsqueda de grabados que plantean en general una mayor dificultad para su observación y reconocimiento. Por otro lado teníamos muy presente que todas estas cavidades ya habían sido estudiadas por prestigiosos investigadores durante casi un siglo y medio.

Con un sistema de iluminación apropiado y la experiencia acumulada por el estudio de numerosos conjuntos de arte paleolítico en la Península Ibérica, centramos nuestra atención en el conjunto de cavidades de Creswell Crags en el Derbyshire. Inmediatamente identificamos numerosos trazos grabados, muy dañados por graffiti, más o menos modernos, sobre todo en Robin Hood Cave y en la cercana cueva de Mother Grundy’s Parlour, donde hemos localizado varias líneas incisas de difícil interpretación y, en el interior de una corta galería, la parte anterior de un caballo grabado en trazo bastante somero. Estos conjuntos están siendo actualmente estudiados en profundidad.

El hallazgo más espectacular es el realizado en la cueva de Church Hole, en el lado sur del conjunto de cavidades de Creswell Crags. Esta cavidad fue excavada en la década de 1870 por el reverendo Magens Mello y Sir William Boyd Dawkins. Los materiales recuperados fueron clasificados por D. Garrod en 1926, como Creswelliense y actualmente se poseen numerosas dataciones radiocarbónicas que sitúan este horizonte cultural entre 12.500 y 12.000 B.P.

En esta cueva hemos identificado por el momento dos paneles decorados. En el primero de ellos se reconoce un macho cabrío dispuesto hacia la izquierda. Se trata de una representación de 57,2 cm. de longitud desde el morro hasta la grupa, por 40,4 cm. de anchura desde la extremidad del cuerno hasta el final de la pata delantera. Aparece en perspectiva semitorcida, es decir están figurados los dos cuernos, pero del resto de los elementos pares, es decir patas, orejas u ojos, únicamente figura uno de ellos. El surco, totalmente patinado, es relativamente ancho y poco profundo.

En el segundo panel, situado a unos 25 metros de la boca, hemos hallado otras dos figuras grabadas que posiblemente representen dos aves en una posición enfrentada. La de la derecha únicamente presenta la zona de la cabeza y posee un cuello muy alargado, motivo por el que pensamos que podría asociarse con una grulliforme o con una anátida (cisne). La silueta de la izquierda, con una cola subtriangular muy alargada y una cabeza que parece presentar un pico curvado, podríamos interpretarla como una rapaz. Las dimensiones de estas representaciones se sitúan en torno a los 30 cm. de anchura por 32 cm. de longitud.

Estilísticamente todas estas figuras se pueden encuadrar en un Estilo IV de André Leroi-Gourhan y por lo tanto pueden fecharse, a falta de un estudio exhaustivo, en la fase final del período glaciar, es decir entre 15.000 y 10.000 años antes del presente.

Nuestro descubrimiento, aún siendo una primicia y a la espera de nuevos hallazgos en breve, coloca por fin a Gran Bretaña en el mapa de distribución del arte rupestre parietal paleolítico. Hasta ahora el ejemplo más septentrional era la cueva de Gouy, cerca de la desembocadura del río Sena en Francia, pero Church Hole está aproximadamente 500 Km. más al norte y esto confirma la gran importancia y potencialidad del conjunto de Creswell Crags. Este hallazgo abre, además, nuevas vías de estudio sobre los modos de vida de unas gentes que habitaron una zona muy próxima a los glaciares perpetuos que cubrían el Norte de Europa. Si la Cueva del Moro en Tarifa (Cadiz) que encontramos en el año 1994 supuso el arte paleolítico más meridional de Europa, ahora en el 2003 Church Hole constituye el arte paleolítico más septentrional del Viejo Continente.

Sergio Ripoll, Paul Bahn y Paul Pettitt



Having just received the above report published on the Internet I consider that some aspects of it need to be discussed and clarified.

Besides the equine engraving on a rib fragment from Robin Hood’s Cave (depicted above) and the anthropomorph from Pin Hole Cave, other palaeoart objects have also been recovered from the Creswell Crags. They include the engraved ivory point from Pin Hole Cave, not mentioned here even though it provides the best comparison with French materials (Armstrong 1925: Fig. 14a-d), two apparently engraved bone fragments from Mother Grundy’s Parlour (Armstrong 1925: Fig. 15-2, Pl. 22-3), and a probable bone pendant fragment from Church Hole Cave, whose chamfered edge is decorated with over a dozen evenly spaced notches (Dawkins 1877: Fig. 6).

The Cresswell Crags palaeoart objects attributed to the very final Upper Palaeolithic have a chequered history. For instance at the time of excavation, in 1875/76, the bone with the horse image was claimed to have been planted, together with a tooth. Dawkins (1877) and Mello (1877) disagreed publicly about this matter and the question of the object’s authenticity remains unresolved. A similar piece, a fake, was later found near Sherborne and published by Smith Woodward (Farrar 1979; Sieveking 1980), who was also duped by the Piltdown finds about the same time. This piece of bone was later dated to 610 BP. Another of the many English Palaeolithic claims that have been rejected concerns the engraved horse mandible from Kendrick Cave, which is now considered to be much younger.

While the Creswellian at about 12 000 years BP is contemporary with the very final Magdalenian, its lithic typology seems to be more closely related to the Tjongerian of Holland and Belgium, the Hamburgian and subsequent Ahrensburgian of Holland and adjacent parts of Germany, and the Brommian of Denmark. These traditions are typologically almost Epipalaeolithic and already herald the appearance of the Mesolithic. So if the few rock art images reported here by Ripoll, Bahn and Pettitt are indeed of the Creswellian, which remains to be demonstrated, they only barely fit the description of Palaeolithic art. While this attribution may sound perfectly reasonable, it needs to be reiterated that, as Bahn himself has argued, co-occurrence at a site with Pleistocene occupation evidence does not make any rock art Palaeolithic. So far the three authors have offered no evidence for the proposed Pleistocene antiquity of these few figures, only opinions based on style as they perceive it.

In stating that Gouy was the northernmost Palaeolithic rock art site up to now, they overlook that in Bahn’s own map of such sites, three are shown more northerly than Gouy: Mladez Cave, which he falsely claims to contain Pleistocene rock art (Bednarik in prep.), and Kapova and Ignatiev Caves in the Urals (Bahn and Vertut 1997: 43). In fact one of these sites is even further north than Creswell Crags, rendering at least the final claim false. This same map also reveals numerous examples of rock art sites Bahn has previously attributed to the Upper Palaeolithic which are in fact younger. This is particularly obvious in the cases of Geißenklösterle and Hohle Fels in Germany, where no rock art has been found at all, despite numerous claims (Bednarik 2002). He also lists three Portuguese sites (Escoural, which is doubtful; and Côa and Mazouco, where no existence of Pleistocene rock art has been demonstrated) and such places as Siega Verde (certainly postdating the Roman period) and the Domingo Garcia group in Spain, which Ripoll also declares to be Palaeolithic. Another example is Ignatiev Cave, where a supposed mammoth motif has recently been dated to 7370 BP (Steelman et al. 2002).

Whilst this eagerness of pronouncing rock art sites as Pleistocene, even those that clearly are not, or possess no rock art all, certainly does not demonstrate that the authors are wrong with their present claim concerning two Creswell Crags sites, their claim will need to be subjected to thorough checking by scientific investigators before it can be considered further. We need to keep in mind that there have been false reports of Pleistocene rock art from Britain before. The earliest case on record was when in 1912 H. Breuil and W. J. Sollas thought they had found cave paintings in Bacon’s Hole, in Wales. It turned out that the red stripes had been made by a workman eighteen years previously. The most spectacular case was that of the ‘Palaeolithic’ rock art found in the Wye Valley, because it found its way into a prestigious journal (Rogers 1981). It is interesting to note that it was the husband of the very same scholar that sealed the fate of the Sherborne fake who eventually debunked the Wye Valley rock art claim (Sieveking 1982).

At this point, an interesting claim has been made that Pleistocene rock art has been discovered in Britain, and it is not the first such claim made. Perhaps this one turns out to be valid, but at this stage there are many questions to be clarified. For instance, why should there be a Palaeolithic depiction of an ibex, a species not present in Britain at the time? Birds are exceedingly rare in authentic Palaeolithic cave art, and there are no swans at all. What is needed now is a thorough investigation of these sites by specialists in micro-morphometry of engravings, modification processes of cave surface phenomena, and dating of engravings. I look forward to seeing this claim tested vigorously. [See postscripts below.]

Robert G. Bednarik
30 June 2003


ARMSTRONG, A. L. 1925. Excavations at Mother Grundy’s Parlour, Cresswell Crags, Derbyshire 1924. Journal of the Royal Anthropological Institute 55: 146-78.
BAHN, P. G. and J. VERTUT 1997. Journey through the Ice Age. Weidenfeld &Nicolson, London.
BEDNARIK, R. G. 2002. Paläolithische Felskunst in Deutschland? Archäologische Informationen 25: 107-17.
BEDNARIK, R. G. in prep. The cave art of Mladec Cave, Czech Republic. Rock Art Research.
DAWKINS, W. B. 1877. On the mammal fauna of the caves of Creswell Crags. Journal of the Geological Society 33: 589-612.
FARRAR, R. A. H. 1979. The Sherborne controversy. Antiquity 53: 211-6.
MELLO, J. M. 1877. The bone caves of Creswell Crags. Journal of the Geological Society 33: 579-88.
ROGERS, T. 1981. Palaeolithic cave art in the Wye Valley. Current Anthropology 22: 601-2.
SIEVEKING, A. 1980. A new look at the Sherborne bone. Nature 283: 719-20.
SIEVEKING, G. de G. 1982. Palaeolithic art and natural rock formations. Current Anthropology 23: 567-9.
STEELMAN, K. L., M. W. ROWE, V. N. SHIROKOV and J. R. SOUTHON 2002. Radiocarbon dates for pictographs in Ignatievskaya cave, Russia: Holocene age for supposed Pleistocene fauna. Antiquity 76: 341-8.

1. Subsequent to writing this, I travelled to England in November 2003 to examine the Creswell Crags engravings together with a local cave art specialist, Professor Kevin Sharpe. He was advised by one of the team, Dr Pettitt, that our visit was not welcome. Readers can draw their own conclusions from this, but it is my impression that the discoverers of Britain’s ‘first Pleistocene cave art’ are not very confident of the veracity of their claim.

2. January 2004. A Spanish reader of this page, Ana María Gomar Barea, has informed us that Ripoll, Bahn and Pettitt have made an error in claiming that they discovered Cueva del Moro at Tarifa (Cadiz) in 1994. That cave was found by the German speleoligist Lothar Bergmann while he was collaborating on a project about the inventory of rock art in the area. For more information visit

3. The first publication on the Church Hole find was submitted a week after the discovery and was accepted by Antiquity the same day it was received, i.e. without refereeing. In it the discovery of three engravings of purported Palaeolithic age was reported, and the main figure described and illustrated as an ibex (see Figure 3a above). The paper offered a long discussion of the significance of an ibex motif from Britain, where no ibex have been reported (actually, ibex did exist in Britain).

4. A second publication followed a year later in INORA. The ‘ibex’ had become a ‘stag’ with now four legs rather than two (see Figure 3b above). The number of Palaeolithic motifs had increased to 47, but most of those depicted appear to be natural features on the cave walls and ceiling onto which the authors had merely projected their interpretations. Again, the report lacked any analytical information, such as data on grooves, micro-topography, weathering or patination, ‘internal analysis’ or archaeometric data.

5. May 2005: a critique by R. G. Bednarik in INORA, calling for proper documentation of the site, elicits an abusive response by Ripoll et al. They present yet another version of the animal figure, this time reverting to two legs and doubling the size of the antlers, as well as presenting other changes (see Figure 4 above). But they fail to address the topic of misidentification of natural features and provide none of the analytical details requested. Their fervent tone (greatly toned down by the editor) only serves to indicate their own uncertainties, and their inability to present proper scientific data. The issue is not whether any of the many markings in Church Hole are of the Creswellian; it is that so far no credible evidence to that effect has been presented, their mode of documentation is inadequate, and their mode of debate is appalling.

6. Subsequently the number of images reported from the site grew to about 90, before growing to 127 motifs, and Church Hole was described as possessing ‘the most richly carved and engraved ceiling in the whole of cave art’, a veritable ‘Sistine Chapel’ of Palaeolithic rock art. However, by 2007, the number of images had dramatically plummeted, by over 90% in fact, with the admission that there were only about 10 motifs now, and only three of them were recognisable. An irrelevant U-Th date of over 13,000 BP had been presented (U-Th results had been rejected over 20 years previously as providing credible rock art dating). This date is from an unsuitably thin calcite skin that is not related to any supposedly Pleistocene rock art. It is, as the following photograph (Figure 5) shows, from a part of the cave not featuring any so-called Palaeolithic engravings. How it should tell us anything about the age of the rock art elsewhere in the cave remains a mystery.


Figure 5. Church Hole, Derbyshire, colour-emphasised image of the cave wall where the speleothem sample subjected to U/Th analysis was removed. There is no presumed Palaeolithic rock art beneath the calcite precipitate.

7. In a paper in the volume Art as metaphor: the prehistoric rock art of Britain, edited by A. Mazel, G. Nash and C. Waddington (Archaeopress, 2007), Bahn and Pettit disagree with Ripoll, whom they now blame for the mistakes in the publications they co-wrote with him in previous years, claiming Church Hole had the richest concentration of Palaeolithic ceiling art in Europe. The background to this development is understood to be that it was actually Ripoll who had discovered the main figures, and when Bahn and Pettitt tried to edge him out of the project, he threatened to sue them. Hence the “break up”.

8. An investigation into the credibility of U-Th dating of thin speleothem skins by Chinese and Australian scientists suggests that this method yields no reliable results (Tang et al. 2020: Rock art of Heilongjiang Province, China. Journal of Archaeological Science: Reports 31; doi:10.1016/j.jasrep.2020.102348). This is demonstrated by analysing split samples as well as by sampling deposits repeatedly. Therefore the U-Th “dating” in Church Hole is not only irrelevant to the age of the nearby rock art, but it also lacks methodological relevance.