Jurasouabe.pdf

A contribution to dating the first development of figurative art in the Swabian Jura © Guy Jouve, 2018. Traduction Paul.G. Bahn Abstact. The portable art discovered in the caves of the Swabian Jura has been dated to the Aurignacian for almost half a century, following work by J. Hahn, but that was not the opinion of the discoverers at the sites of Vogelherd and Hohlenstein Stadel. The dating of these figurines poses questions about the first development of figurative art. This paper examines the validity of the arguments presented about radiometric ages of the finds comparing to their stratigraphic locations. A varied chronology for these artworks becomes clear: in our view, some pieces from Vogelherd and Hohle Fels date to the Gravettian, while others from Vogelherd and Hohlenstein Stadel date to the Magdalenian. It seems legitimate to recognize some of these artworks as the first developments of figurative art in the earliest Mid Upper Paleolithic/Gravettian. The arguments in favour of the Aurignacian do not hold up to critical examination. Introduction The beginnings of figurative art mark a major milestone in the evolution of human behavior, and discovering that moment is a task for the archaeologist. It is known that the Aurignacians in Southwest France depicted sexual symbols, but the development of figurative art needs to be sought elsewhere. One region has been a candidate for this since the late 196os: the Swabian Jura, with its figurines, based on the interpretations of J. Hahn. A few decades later, J. Clottes proposed similar dates for the parietal art of Chauvet Cave, which might support the idea that an evolved figurative art developed in that period in two regions of Europe (Clottes et al. 1995). These conclusions stand or fall on the reliability of the dating methods and techniques used. It has never been possible to prove that the art of Chauvet Cave dates to the Aurignacian (Combier, Jouve 2012; Pettitt, Bahn 2015; Jouve 2018); but where the sculptures of the Swabian Jura are concerned, it is necessary to tackle the problems posed by their dating, starting with the observations of these discoveries in the field. Were the Swabian Alps an artistic desert during the Gravettian and Magdalenian? While Aurignacian art is famous at four caves in this region, the Gravettian and Magdalenian did not, apparently, yield any portable artworks according to the best known publications; this is the precise opposite of what is seen elsewhere in Europe where the art is poor or even non-existent in the Aurignacian but arises in the Gravettian and flourishes in the Magdalenian. An examination of the productions at the sites of Vogelherd, Hohle Fels, Brillenhöhle, Weinberghöhlen, Felsställe, Hohlenstein Stadel and Hohlenstein bei Ederheim leads us to note that the evolution of human capabilities continued after the Aurignacian in Swabia until the Magdalenian, in artistic production as well as in tool industries. 1 1. Is there Gravettian portable art in the Swabian Jura? Several sites have revealed human occupations in the Gravettian : SteinackerMüllheim, Hohle Fels, Girgenstein, Geissenlkösterle, Brillenhöhle, Bockstein-Törle, Vogelherd, Weinberghöhle, Abri in Dorf, Mittlere Klause, Salching (Hahn 2000). 1.1. Riek’s excavations at Vogelherd The radiometric dating of bones discovered in the cave of Vogelherd has yielded more Gravettian dates than Aurignacian, distributed throughout all levels (table 1) : 8 Gravettian dates, 6 Aurignacian dates, 2 Magdalenian dates, 3 non-determined dates (at the Aurignacian/Gravettian boundary); but also some dates from the end of the Neolithic in V and IV, from human bones. Conard et alii (2003) specify: « The collagen values of the faunal remains from Vogelherd are generally high and indicate that the dates are reliable. The sample preparation of the specimens should in general have been adequate to eliminate any significant contamination ». The presence of micro-gravettes in layer IV confirms the Gravettian. The range of dates in each level shows that there has been considerable mixing of sediments. Eleven statuettes were discovered in layers IV and V which Riek attributed to different cultures -- layer IV to the Upper Aurignacian and layer V to the Middle Aurignacian, according to the nomenclature in use when the excavations were carried out in 1931. “We do not find any precise indication of their stratigraphic position, but one thing is noteworthy: in the large monograph on Vogelherd, we only learn that Riek considers the fully rounded figures as belonging to the Middle Aurignacian, and the half-reliefs as Upper Aurignacian” (Freund 1957: 16). So there is an orderly vertical deposit of figurines. The cross-shaped striations engraved on several sculptures in both layers could show that all these objects come from the same period (figure1a). As one does not pile up objects above the ground, this must be a cache dug close to the boundary between layers IV and V, and hence contemporaneous with upper layer IV, which is principally Upper Aurignacian. It is regrettable that we do not know anything of the circumstances of such an important discovery. Layer IV Animal Layer V Horse Fig.1a. Cultural interpretations We must also bear in mind the evolution of the nomenclature. At the time of the discovery, « The Châtelperron [culture] became the Lower Aurignacian, and that of La Gravette, the Upper Aurignacian, with the adjective Middle reserved for that of Aurignac, because, wherever these three cultures were found together, they seemed to 2 form a succession from the bottom upwards […]. This succession has been maintained until the last few years without any new observations modifying it » (Peyrony 1933). Since the 1960s, we have used a different system: the Upper Aurignacian is called Gravettian, and so layer IV belongs to the Gravettian according to Riek, as do the statuettes too. Riek noted analogies with the statuettes of Pollau, Predmosti (figure b), and the Pavlovian of Kostenki, which are all contemporaneous with the Gravettian. This was confirmed by Leroi-Gourhan who clearly assigned the figurines from Vogelherd Cave to his style II, which is chronologically equivalent to the Gravettian (LeroiGourhan 1965: 63, 67-68, 244-245). For Delporte (1993), the anthropomorphic statuette could be associated with the Gravette points and contemporaneous with Gravettian statuettes. Freund (1957) reckoned that the Aurignacian habitation had begun later than Riek had indicated. Predmosti, Gravettian, Brno Museum. Fig.1b. Vogelherd, layer V ( photo Don Hitchcock). With the exception of these authors, the statuettes are more or less universally attributed to the Aurignacian in present-day terminology. This began with de Sonneville Bordes (1965) and Hahn (1970). In evaluating the statistical distribution of the lithic industry discovered in layers IV and V, using the method developed by Bordes at Bordeaux, where Hahn was a pupil, they settled on the same culture – Aurignacian in the current sense – for both levels IV and V, which was in contradiction with Riek for layer IV. Since this Aurignacian differs from that found in France, they called it Aurignacian of Germanic. They were unaware of the numerous Gravettian dates and, when Hahn learned of them, he extended the Aurignacian to 23,000 BP (Hahn 1995). Their method was founded on a statistical average of the lithic industry, with no knowledge at all of the location of the archaeological pieces in the layers, since Riek gave no indication of this (the only location reported is that of a split-base point in IV, close to the base, but de Sonneville Bordes took no account of the proximity of V). In particular, this method ignored the mixing of sediments which had taken place, for example « a mix of Upper Aurignacian amid the Middle Aurignacian layer » (Riek 3 1933). These mixings were confirmed by the radiometric dates (table 1): in layer III the dates obtained are Gravettian and Aurignacian; in layer IV they are Gravettian, Aurignacian and Magdalenian; in level V they are Gravettian and Aurignacian (through comparison with the uncalibrated dates currently assigned to these cultures). Finally, the attribution by de Sonneville Bordes and the young Hahn was based on the knowledge of the 1960s concerning the cultural attribution of the industry; as a result of new observations, specialists now have a different analysis. Moreover, Riek noted resemblances between the Gravettian assemblage of Brillenhöhle and the Aurignacian of Vogelherd. « The rich assemblage from Brillenhöhle showed in Riek’s view affinities with the Aurignacian from his excavation at Vogelherd » (Conard, Moreau 2004; Riek 1973). Many stone tools are also present in the Gravettian layers as well as in the Aurignacian layers of the sites in the Swabian Alps. The two split-base points from Vogelherd are certainly characteristic of the Aurignacian, but they are only present in layer V and at the base of layer IV. Shortly before his death, Hahn (2000: 254) wrote in a posthumous publication that layer IV at Vogelherd contained some Gravettian. So the arguments he presented in his youth in favour of the Aurignacian are obsolete, and the Gravettian must be retained. Since the initial Gravettian is absent from the site according to Hahn, the statuettes might come from a later Gravettian, around 23,000 BP, which is very understandable in terms of the extremely developed artistic qualities of the bestpreserved figurines. Table 1. Uncalibrated radiocarbon dates, on bones from Vogelherd. A : Aurignacian ; G : Gravettian ; M : Magdalenian (late). (After Conard & Bolus 2003; Hahn, 1977). Layer III Ref. OxA-10196 Conv/AMS Years BP SMA 25 780 ± mammoth tooth 250 Culture G dentin (root) OxA-10198 SMA 26 110 giant deer tooth 310 G 310 A 650 ? 55 M dentin (root) OxA-10195 SMA 31 680 mammoth tooth dentin (root) OxA-10197 SMA 39 700 woolly rhino. tooth dentin (root) IV KIA 8966 SMA 13 015 bovid/horse femur frag IV/V KIA 8957 H-4053-3211 PL0001340A SMA Conv SMA 26 160 longbone frag 150 30 730 mixed bone sample 750 13 630 reindeer 410 G A/G M metatarsal GrN-6583 Conv 23 860 mixed bone 190 G 27 630 burned bone 32 180 horse tibia 34 100 bovid/horse rib 23 020 bone-Mammoth? 830 960 1100 400 G A A G sample V GrN-6662 PL0001339A PL0001342A H-4035-3209 Conv SMA SMA Conv 4 H-8498-8950 H-8497-8930 GrN-6661 H-8499-8991 H-4056-3208 H-4054-3210 H-8500-8992 KIA 8968 PL0001338A KIA 8969 KIA 8970 PL0001337A Conv Conv Conv Conv Conv Conv Conv SMA SMA SMA SMA SMA 25 900 mixed bones 27 200 mixed bones 30 650 burned bone 31 350 mixed bones 31 900 mixed bones 30 162 mixed bones 30 600 mixed bones 31 790 artiodactyl tibia 32 400 horse tibia 32 500 reindeer bone frag 33 080 horse longbone 35 810 longbone frag. 260 400 560 1120 1100 1340 1700 240 1 700 260/250 320/310 710 G G A A A G/A G/A A A A A A Today, the attribution of layer IV of Vogelherd to the Aurignacian is presented as coming directly from Riek: Upper Aurignacian = Aurignacian (Conard et al. 2003: table 1).This conceals the fact that Upper Aurignacian (Riek) = Gravettian according to the change in nomenclature around 1960 for the European Palaeolithic. but Upper Aurignacian (Riek) ? = Aurignacian non-according to the change in nomenclature around 1960 , this is an archaeological reattribution initiated by de Sonneville Bordes and Hahn for layer IV of Vogelherd at the end of the 1960s. One simply cannot maintain the Aurignacian of layer IV if, at the same time, one acknowledges for Riek that « there is every reason to assume that the majority of these designations are correct » (Conard et al. 2003: 80). 1.2. Hohle Fels (Schelklingen) Fig.2. Hohle Fels, in the circle, place where figurines were found. (After Hahn 2000) 5 Discovery and dates Several statuettes carved in ivory have been discovered since 1999 in a recess 20 metres from the entrance of the cave of Hohle Fels (figure 2) located in the Ach Valley. The stratigraphic study of this zone was carried out in a far more detailed way than at Vogelherd. Some post-depositional displacements are clear: Hahn (2000) identified a movement of Gravettian sediments from the inside of the cave to the outside – through ursine bioturbation and cryoturbation. This is confirmed in particular in unit D through the inclination of the geological levels visible in the stratigraphic sections (figure 3), both East-West and also North-South; through an assemblage of radiometric dates belonging to cultures from the Aurignacian to the Gravettian (table 2); through the presence of rounded gravel in the sediments in AH IIc, IId, IIe (Schiegl et al. 2003), in AH III (Goldberg et al. 2003); and finally through the positions of the pieces of two statuettes (horsehead, and bird) which are at some distance in the direction of the sediment displacement (figure 3). Hohle Fels. Schematic plan of the cave with excavation area and East–West Profile. A: Magdalenian, C: Gravettian, D: Gravettian –Aurignacian, E: Aurignacian 4 (in Floss & Kieselbach 2004). 6 Fig.3. Positions of the statuettes and cultural attribution by Conard 2003: Unit D attributed to Aurignacian. Unit C groups together layers with a purely Gravettian content. Among the artifacts are: • Sandstone sculpture of a phallus in IIcf discovered in 2005, Gravettian level dating to 27,000 – 28,000 BP. Fig.4. Polished and engraved reindeer antler (after drawing C. Pasda, in Scheer 2000) • Decorated reindeer antler extracted from level IIc (figure 4); this is the only figurative artifact in the whole Swabian Jura that has been dated directly: Oxa 4599 28 920 ± 440 BP Oxa 5007 29 550 ± 650 BP (Common portion 29 120 ± 360 BP) 7 On this animal drawing, some parallel segments are engraved, as on other Gravettian artifacts from the site. Another reindeer antler has rhombohedric parallel striations engraved on it. One also finds these parallel striations on certain sculptures from Hohle Fels: horsehead, and Venus (figure 5). This kind of striations also exists on certain figurines from Vogelherd. Unit D, the levels are seriously disturbed. The statuettes were discovered in these levels, from IId to V, a zone of constricted and jumbled sediments (figure 3). Horse head. (After Cook, J.,1) • The sculpture of an animal head often identified as a horse was discovered in 1999 following excavations carried out under the direction of Conard and Uerpmann in a Gravettian level (Conard, Floss 2000). This is geological level GH3d, also called AH2d. «The lithic assemblages from these deposits [IId IIe]are relatively poor in artifacts, but based on the presence of small numbers of microgravettes2, carinated burins and a nosed end scraper3, [....] the raw material spectrum falls within the range of Gravettian assemblages » (Conard, Moreau 2004). One finds here a change in the indication of the position at the time of discovery: from IId to IIe (Conard & Floss 2001: 242) to GH3d (= IId), then on the next page it is an intermediary position between GH 3 and GH5, that is, between IId and IIe. Then, in the section, it is practically in IIe (figure 3). Interpretations Next – and this is doubtless the objective that was sought – a revision of the cultural attribution of the levels is performed. Conard marks IId and IIe as Aurignacian on the stratigraphic section. Why? The (fake) explanation is provided: « the organic assemblage includes an ivory figurine which resembles the head of a horse and is a typical Aurignacian4 find » (Conard & Moreau 2004). This lack of respect for the initial indication by the discoverers is unexpected. The statuette was definitely discovered in level IId. In view of that layer’s thinness, one might at best have written that it was in an area where it was hard to distinguish 1 Ice Age art: arrival of the modern mind. The British Museum, 18 Feb 2013). 2 Gravettian Aurignacian 4 See figures 1b ; 5. 3 8 between sediments IIc IId and IIe. In any case, Gravettian flints are present. As for the radiometric dates (table 2), they do not enable one to distinguish the end of the Aurignacian from the start of the Gravettian. It is therefore unsuitable to deduce from this that the statuette was discovered in an Aurignacian level. The real question here is knowing which level the statuette was in when it was first abandoned. One piece of useful information is provided by the discovery of a fragment of the cheek of the horsehead in a lower level, IIIa, more than a metre away from the rest of the head. How could one imagine that the dynamic pressure of the sediments could have pulled away a fragment of the statuette and sent it more than 1m from the rest of the object, without displacing the original object, even by a few centimetres? This would defy all laws of mechanics. So one needs to accept that the statuette was originally in a level higher than II d, perhaps in IIc, and in any case in a Gravettian level (the thickness of level IId is about 10 cm at this spot). Three statuettes Two statuettes, depicting a bird and a human has been discovered in III and IV, where gravettian dates were found (table 2). Later a woman (Venus), as well as a fragment of another Venus, have been discovered (figure 5). Fig.5. Left: Venus of Hohle Fels . Right: Dolni Vestonice V (Gravettian) photos Silosarg, D. Hitchkock Close to the fragments of this Venus discovered in six pieces, there were limestone blocks, several decimetres in size. The layers of sediments must have been broken by the fall of these blocks from several meters, and the statuette fragmented by the gel. Who could possibly guarantee that the statuettes remained within the same sediment? Dates in the level: 31 000 to 35000 BP; 41 000 BP (50 cm away). « In Russia, in loess sites (Avdeevo, Kostenki etc.) Venuses are often hidden in visible cavities, in which case the containing level is older than its content. The attribution of this female statue of Hohle Fels, alone to the Aurignacian throughout Europe from the Urals to the Atlantic, would be an anachronism; the Venus are the most original cultural trait of the Gravettian » (Combier pers.com. 2018). 9 Table 2. Dates uncalibrated, BP, Hohle Fels (taken from Conard, Bolus 2006: table 3) 1.3. Brillenhöhle : Gravettian Vénus Near Hohle Fels, Riek (1973) found a Venus (1955-1963), broken, in a Gravettian layer. He failed to rebuild it, which was “his bitterest disappointment”. Brillenhöhle VII : KIA-19549 (Mammoth)/Rhino. rib KIA-19553 (Mammouth)/Rhino. rib B-492 burnt bone Fragment, ivory, 3cm x 2,5cm 27 030 ± 180 BP 25 870 ± 230 BP 25 000 BP (Photo Landesmuseum Württemberg, Stuttgart). 10 1.4. Geissenklösterle cave The results of dating in the cave of Geissenklösterle raise some difficulties which we do not think have yet been solved. Located in the Ach valley, a few kilometres downstream of Hohle Fels, the cave of Geissenklösterle yielded several statuettes carved in ivory, during excavations carried out by G.A. Wagner and then J. Hahn between 1973 and 1991. The recorded geological levels GH 1 to 17 were interpreted as archaeological units from I to V with subdivisions (IIn, IIa, IIb...). The lines separating the levels are more or less horizontal; there are stone blocks from different collapses of the wall and ceiling, especially in level II from which the statuettes were extracted. Hahn (1988) noted that the reconstitution of objects revealed that the initial location of the artefacts and bones had changed through cryoturbation and bioturbation after the first sedimentation, and that the direction of displacement of the artifacts was primarily vertical, extending through several geological horizons. He also emphasised that some apparent mixing was the result of the difficulty of reliably defining the stratigraphic units within the deposits which contained an abundance of large limestone blocks, often of bigger size than the thickness of the archaeological subdivisions. He subsequently abandoned the subdivisions. Level II which yielded the statuettes This level is located below level I attributed to the Gravettian by the coherence of its lithic contents. This level II (GH 11 12 13 14) begins just below GH 10 which revealed a cryoturbation. In an area with an abundance of rubble, the sediments contain more clay than the lower levels (Richter et al. 2000). The sediments are of a different colour from those of the upper levels, browner, but the rubble also seems more corroded.The upper zone IIn seems to have been affected by solifluction, as shown by crushed flints and rolled and polished bones. Its tools are too few and too crushed to allow definite attribution (Hahn 1982). The radiometric dates were obtained through several techniques, radioactive counting then AMS measurement, and several processes of decontamination, filtration and ultrafiltration. The results sometimes show differences of several millennia in the same level, and their comparison provides information. The first dates were obtained from (conventional) radioactive counting on mixtures of poorly identifiable bones, and bones used by humans were not specially selected. The dates obtained for level II (called Aurignacian) range from 26,650 to 35,450 BP (Richter et al. 2000), which confirms a major disturbance of the sediments by bears and burrowing animals and cryoturbation. The dating of burned bones failed. So are Aurignacian and Gravettian dates in level I called Gravettian. Then 14 bones selected as having been related to human activity were dated by AMS; each yielded two dates (table 3), one obtained through ultrafiltration the other without ultrafiltration (Higham et al. 2012). One can observe that in a Gravettian layer Ic, the Aurignacian date obtained without ultrafiltration is younger than that obtained with ultrafiltration, but the Gravettian date obtained in Gravettian layer It remains 11 practically identical in both methods. All the other dates, in layers II and III, become older after ultrafiltration. How to interpret the differences in dates? Ultrafiltration makes it possible to sort molecules by size, and eliminate the smallest. The comparison of the results shows that the dates obtained after ultrafiltration are in most cases more than a thousand years older, so one can simply deduce that the eliminated small molecules are clearly the youngest and that each sample dated without ultrafiltration contains several substances of different ages. Our aim is to know not the age of the bones but the age distribution of the substances present in layer II which yielded the statuettes; knowing whether or not the youngest molecules are contaminants is a secondary concern: the dates show that there is material of Aurignacian age, but also younger material in the level containing the statuettes. One cannot attribute the statuettes to one of these periods without providing some justification. One should note that these dates were only obtained from bones, none was obtained from wood charcoal, and none from organic material in the sediments, which has doubtless resulted in partial information. One statuette depicting a mammoth was fond broken into a hundred fragments in a zone use as a passage by animals (Hahn 1982). So one would expect that the debris would be somewhat dispersed, but in fact they were scattered over 20 cm 2, which is very limited. What prevented dispersal? The explanation could be that the statuette was imprisoned in a hole. So one must envisage the presence of a hollow. A hole, 20 cm deep, dug in the Gravettian, and then filled by the passing of animals and by rockfalls, would reach an Aurignacian level. Mammoth (Photo: Don Hitchcock 2015) But if the statuettes belonged to the Aurignacian, along with the lithic and organic remains in level II, that would mean they were not placed in holes. But one would then need to find a different explanation for the origin of the younger molecules which are only eliminated by ultrafiltration: contamination during excavation or in the laboratories? « Only after clarifying these issues would it be possible to test the models available for explaining the timing of the advent of innovations attributed to the Aurignacian » (Higham et al. 2012). 12 Taking into account all the information at our disposal, we believe that the radiometric dates obtained do not make it possible at present to assign the Geissenklösterle statuettes to a specific culture. Table 3. Comparisons of radiocarbon dates from bone samples. Extraction with or without ultrafiltration (±) BP uncalibrated. Dates carried out by ORAU except k : KIA. (After Higham et al. 2012). Level Exchange of ions-gelatinisation Gelatinisation Ultrafiltration = It Ic IIa 27 950 ± 550 30 300 ± 750 33 200 ± 800 27 960 ± 290 32 900 ± 450 33 000 ± 500 IIb 29 800 ± 240 (k) 31 870 +260/-250 (k) 33 950 ± 550 34 800 ±600 III 30 100 ± 550 31 180 +270/-260 (k) 32 900 ± 850 (k) 40 200 ± 1600 35 050 ± 600 37 400 ± 800 36 850 ± 750 38 900 ± 1000 IIIa 34 800 +290/-280 (k) 34 330 +310/-300 36 650 ±750 36 850 ± 800 IIIb 34 080 +300/-290 (k) 34 220 +310/-300 (k) 28 640 +380/-360 (k) 36 100 ± 700 37 800 ± 900 37 300 ± 800 IIIc 33 600 ± 1 900 32 050 ± 600 39 400 ± 1 100 38 300 ± 900 = Not significant 8,1 mg of collagen extracted, the most reliable value of the whole series of measurements by ultrafiltration. 2. Magdalenian art in the Swabian Jura In the region of the Upper Danube, the number density of Magdalenian sites is quite remarkable: 13 Distribution of Magda1enian sites . (1) Moosbühl; (2-12) Rislisberghöhle, Käsloch, Köpfli, Koh1erhöh1e, Brügglihöhle, Heidenküche, Bütten1och, Birseck-Ermitage, Kastelhöhle, Abri Chesselgraben, Hollenberghöhle 3; (13-15) Schweizersbild, Kesslerloch, Freudenthal; (16) Munzingen; (17) Teufelsküchen; (18, 19) Petersfe1s, Gnirshöh1e; (20) Bildstockfels; (21) Probstfe1s; (22) Napoleonskopf; (23) Dietfurt; (24) Schussenque!le; (25-32) Felsställe, Hohle Fels Hütten, Schmiechenfels, Hohler Fels Schelk1ingen/Helga-Abri, Sirgenstein, Geissenklösterle, Brillenhöhle;(33) Burkhardtshöhle; (34-37) Spitzbubenhöhle, Vogelherd, HohlensteinStadel/Hohlenstein Kleine Scheuer; (38) Kleine Scheuer Rosenstein; (39-40) Bärenfelsgrotte/Spitalfels; (41-43) Kleine Ofnet, Hohlenstein Ederheim, Kaufertsberg; (44-46) Kastlhänghöhle, Klausenhöhlen, Heidenstein; (47) Barbing; (48) Steinbergwand; (Weniger 1989). Female depictions of the Lalinde-Gönnersdorf type are distributed in several sites of this region. 2.1. Vogelherd Excavations in the sediments of Riek’s spoilheap. N. Conard and M. Zeidl carried out excavations from 2005 to 2014 in the sediments thrown outside by Riek’s team in 1931. Numerous objects were discovered, including a Venus of Lalinde-Gönnersdorf type (figure 6) carved in a boar tooth5 and heavily polished (Südwest Presse 2010); it seems compatible with the two Magdalenian dates of ca. 13,000 BP from layer IV which yielded several statuettes. Blaubeuren Museum compared it (Spanel 2014) with the Venuses of the same type discovered at Petersfels near Lake Constance during the work by Peters, Toepfer, Mauser and Albrecht, carved from lignite, and dated to about 14,000 BP. « However, the figures of Lalinde Gönnersdorf highlight the posterior part of the body and not the belly. These advanced stylizations evoke a much later period than naturalistic representations » (Combier pers.com. 2018). So are some female engravings found in La Marche. 5 https://www.urmu.de/de/Forschung+Arch%C3%A4ologie/Eiszeitkunst/Vogelherd/WildschweinVenus-(Nachgrabung) 14 Fig.6. Vogelherd. Venus carved and polished on boar tooth; 6,9 cm. (Photo Thilo Parg, at Blaubeuren Museum) 2.2. Hohlenstein bei Ederheim In 1912, six Venuses engraved on a limestone slab were discovered at Hohlenstein bei Ederheim by F. Birkner (Bosinski 1982) (figure 7). Fig.7 Engravings. Hohlenstein bei Ederheim (after Bosinski 1982, Bosinski& Fischer 1980) On the slab which bears the engraved Venuses, a horse’s head and its legs are also depicted. This association can be compared with an engraved slab from Gönnersdorf bearing drawings of several Venuses with a horse. 2.3. Hohlenstein Stadel Excavations by Wetzel and Völzing The famous statuette known as the Löwenmensch, reconstituted from numerous fragments (figure 8), came from a Magdalenian level according to Wetzel and Völzing who directed the excavations: « … the Magdalenian yielded a long bone needle, two perforated pendants… and fragments of an ivory sculpture. What it depicts is not recognizable; the sculpture is perfectly intentional» (Letter from R. Wetzel to H. Schlief, 26/08/1939, in Wehrberger 2013: 14). Fig.8. Restauration 2013 (photo Silosarg). 15 Contradictions Thirty years later, Hahn contested the stratigraphic location of the discovery, and estimated that it had been found in an Aurignacian layer (Hahn 1970, 1971a, 1971b). He used two arguments: 1- The indication of depth written on the box, in bad condition, which he never copied. According to him, it corresponded to a depth of 1 to 1.2 m. 2- The reddish-brown colour of the statuette’s ivory, which he claimed could not come from the Magdalenian, because the site’s Magdalenian bones are yellow. But they are not ivory! He also quoted Wetzel who indicated the yellow and reddishyellow colour of the sediments in the Aurignacian (Hahn 1971b: 12), but reddishbrown at a Magdalenian depth according to a section by Völzing which he also cited (Hahn 1971a: 235); all of this destroys his conclusions on the colour of the statuette’s fragments. He wrote that the statuette cannot be Gravettian, because no Gravettian was known in the valley. But it is now known that there is indeed some Gravettian in the Lone Valley, for example at Bockstein-Törle; and moreover Hahn himself later recognized that the Gravettian is present at Hohlentein Stadel IV (Hahn 2000). So the arguments that he published in his youth are totally obsolete. To contest the report by the discoverers on the position of the statuette requires proofs which have never been presented. More recently, excavations have been carried out in order to date the 6th spit in which the statuette fragments were supposedly discovered (Kind et al. 2014). The dating of the prolongation of this level shows that it is indeed Aurignacian, which leads the authors to accept that the statuette is Aurignacian. But they bring no justification for the statuette’s position in this layer other than this claim: « The find box with the fragments is clearly labelled ». Moreover, Hahn and his successors have never pointed out that their opinion contradicted that of the discoverers, and they have never produced a photograph of the box or a copy of its inscription. Oowever, a bone taken from the box has given the radiocarbon date 12 400 ± 180 BP, Magdalenian, (Schmid et al. 1989: 94). No valid reason has ever been presented to contest the excavator’s report that the statuette was discovered in a Magdalenian level. 2.4. Magdalenian female depictions in other sites In the cave of Hohle Fels (Schelklingen): a piece of carved lignite discovered in the Magdalenian could be a fragment of a Venus according to Hahn (Weniger 1989: 358): “a still-unpublished fragment ofjet which might be the back of a Venus (J. Hahn, personal communication)”. Badly preserved red dots on a fragment of limestone have also been discovered. In the rock shelter of Felsställe: a schematic engraved Venus was discovered in Magdalenian rubble (Haas-Campen 1997; Kind 1987). 16 Conclusions The dating of figurines is a particularly delicate problem, because it is doubly indirect: one dates bones or bits of charcoal contained in the strata in order to deduce an age for each layer, and then one assimilates the age of each statuette to that of the layer that contains it. Stratigraphic readings with a resolution of a few centimeters or even a decimetre in disturbed areas have poor value. In order to strengthen the reliability of such an approach, it would also be useful to date organic material of the sediments, which is now possible with AMS using samples of a few dozen grams; it might also be useful to carry out magnetic measurements that could provide information on the directions in which sediments have moved. And if there is sufficient collagen present, why not envisage direct dating of ivory fragments, since modern techniques make it possible to safeguard in 3D all the physical characteristics of the fragment that is to be lost? Even though there is a broad consensus that places the figurines of the Swabian Jura in the Aurignacian, this should not prevent the investigation of other solutions to the problem of their dating, which in our opinion has by no means been solved. The documents we have examined do not allow us to contradict the interpretations of the discoverers on the pretext of the early nature of their work and the success enjoyed by their successors and contradictors. It is far from well-established that all these statuettes are of the same period; and it is probable that those of Hohle Fels and some of those from Vogelherd are indeed at the origins of figurative art, but in our view they are Gravettian and not Aurignacian. It appears quite certain that the region of the upper Danube has yielded artworks of the Gravettian and Magdalenian, like many other parts of Europe in these same periods. Acknowledgements The author would like to thank Paul Bahn for translating this text; and Jean Combier, João Zilhão and Paul Bahn for help with documentation and comments. It goes without saying that the opinions in this paper are mine alone. References BOSINSKI, G. 1982: Die Kunst der Eiszeit in Deutschland und in der Schweiz. Kataloge Vor- und Frühgeschichtlicher Altertümer, Val. 20, Rudolf Habelt, Bonn. BOSINSKI, G. 2013: Les précurseurs de l'art aurignacien. In: Le Paléolithique supérieur ancien de l'Europe du Nord-Ouest”. Mémoire LVI de la Société Préhistorique française, 497-511. BOSINSKI, G. & FISCHER, G. (1980). Mammut- und Pferdedarstellungen von Gönnersdorf. Franz Steiner, Wiesbaden. 17 CLOTTES, J., CHAUVET, J.-M., BRUNEL-DESCHAMPS, E., HILLAIRE, C., DAUGAS, J.-P., ARNOLD, M., CACHIER, H., EVIN, M., FORTIN, P., OBERLIN, C., TISNERAT, N. & VALLADAS, H. 1995: Les peintures paléolithiques de la grotte Chauvet”. Comptes-rendus de l’Académie de Sciences 320 II a: 1133-1140. COMBIER, J., JOUVE, G. 2012: Chauvet cave’s art is not Aurignacian: a new examination of the archaeological evidence and dating procedures”. Quartär 59: 131–52. COMBIER, J., JOUVE, G. 2014 : Nouvelles recherches sur l’identité culturelle et stylistique de la grotte Chauvet et sur sa datation par la méthode du 14C. L’Anthropologie. Volume 118, Issue 2, Pages 115-210. CONARD, N.J. 2003: Palaeolithic ivory sculptures from southwestern Germany and the origins of figurative art. Nature, Vol. 426:830. CONARD, N. J., FLOSS, H. 2000: Eine Elfenbeinplastik des Hohle Fels bei Schelklingen und ihre Bedeutung für die Entwicklung des Jungpaläolithikums in Südwestdeutschland. Archäologisches Korrespondenzblatt 30, 4: 473-480. CONARD, N.J., NIVEN, L.B., STUART, A.J. 2003: The Chronostratigraphy of the Upper Paleolithic Deposits at Vogelherd. Mitteilungen der Gesellschaft für Urgeschichte – 12:73-86. CONARD, N.J., BOLUS, M. 2003: Radiocarbon dating the appearance of modern humans and timing of cultural innovations in Europe: new results and new challenges. Journal of Human Evol. 44: 331– 371. CONARD, N.J. & BOLUS, M. 2008: Radiocarbon dating the late Middle Paleolithic and the Aurignacian of the Swabian Jura. Journal of Human Evolution 55: 886–897. CONARD, N.J. & MOREAU, L. 2004: Current Research on the Gravettian of the Swabian Jura. Mitteilungen der Gesellschaft für Urgeschichte: 29-59. DELPORTE, H.1993: L'image de la femme dans l'art préhistorique. Nouvelle édition augmentée. Paris. Picard. FLOSS, H., KIESELBACH, P. 2004; The Danube Corridor after 29,000 BP – New results on raw material procurement patterns in the Gravettian of southwestern Germany. Mitteilungen der Gesellschaft für Urgeschichte – 13: 61-78. FREUND, G.1957: L’art aurignacien en Europe centrale. Préhistoire et Spéléologie ariégeoise. Bull. de la Soc. Préhist. de l'Ariège. Vol. 12: 1-24. GOLDBERG, P., SCHIEGL, S., MELIGNE, K., DAYTON, C., CONARD, N. 2003: Micromorphology and Site Formation at Hohle Fels Cave, Swabian Jura, Germany. Eiszeitalter und Gegenwart. Hannover, 53, 1-25. HAAS-CAMPEN, S. 1997: Ulm und der Alb-Donau-Kreis. Stuttgart 1997: 131-133. HAHN, J. 1970: Die Stellung der männlichen Statuette aus dem Hohlenstein-Stadel in der paläolithischen Kunst. Germania 48, 1-12. HAHN, J. 1971a). “La Statuette masculine de la grotte du Hohlenstein-Stadel (Wurtemberg)”. L’Anthropologie. Tome 75, n° 3-4: 233-234. 18 HAHN, J. 1971 b: Eine jungpaläolithische Elfenbeinplastik aus dem Hohlenstein-Stadel. 565 Fundberichte aus Schwaben. N.F. 19: 11-23. HAHN, J. 1977: Aurignacien, das ältere Jungpaläolithikum in Mittel- und Osteuropa. Köln/Wien: Böhlau-Verlag. HAHN, J. 1982: Demi-relief aurignacien en ivoire de la grotte Geissenklösterle, près d'Ulm (Allemagne Fédérale). Bulletin de la Société préhistorique française. 1982, tome 79, N. 3: 73-77. HAHN, J. 1988: Die Geißenklösterle-Höhle im Achtal bei Blaubeuren I. Fundhorizontbildung und Besiedlung im Mittelpaläolithikum und im Aurignacian. Stuttgart: Theiss. HAHN, J. 1995: Neue Beschleuniger-14C-Daten zum Jungpaläolithikum in Südwestdeutschland. Eiszeitalter u. Gegenwart.45: 86-92. HAHN, J. 2000: The Gravettian in Southwest Germany – environment and economy. In: Roebroeks, W., Mussi, M., Svoboda, J. & Fennema, K. (eds), Hunters of the Golden Age. The Mid Upper Palaeolithic of Eurasia 30,000 – 20,000 BP. Analecta Praehistorica Leidensia 31. Leiden: University of Leiden: 257- 270. HIGHAM, T., BASELL, L., JACOBI, R., WOOD,R., BRONK RAMSEY,C., CONARD, N.J. 2012: Testing models for the beginnings of the Aurignacian and the advent of figurative art and music: The radiocarbon chronology of Geißenklösterle. Journal of human evolution. Volume 62. Issue 6: 664-676. HOUSLEY, R.A., GAMBLE, C.S., STREET, M., PETTITT, P. 1997: Radiocarbon evidence for the late glacial human recolonisation of Northern Europe. Proc. Prehist. Soc. 63: 25–54. JOUVE, G., 2018. Exemple d'utilisation de l'isotope 13C dans la recherche de l'origine du carbone : grotte Chauvet. In Press. DOI: 10.13140/RG.2.2.29915.4688. KIND, C.J. 1987: Das Felsställe. Eine jungpaläolithisch-frühmesolithische Abri-Station bei EhingenMühlen,Alb-Donau-Kreis. Forschungen und Berichte zur Vor- und Frühgeschichte BadenWürttemberg. Konrad Theiss Verlag, Stuttgart. KIND, C.J. EBINGER-RIST, N., WOLF, S., BEUTELSPACHER, T. , WEHRBERGER, K. 2014: The smile of the Lion Man. Recent excavations in Stadel Cave(Baden-Württemberg, south-western Germany) and the restoration of the famous Upper Palaeolithic figurine”. Quartär 61: 129–145. LEROI-GOURHAN, A. 1965: Préhistoire de l'art occidental. Paris, Mazenod. PETTITT, P., BAHN, P. 2015: An alternative chronology for the art of Chauvet cave. Antiquity 89, 345, pp 542–553. PEYRONY, D. 1933: Les Industries « aurignaciennes » dans le bassin de la Vézère. Bulletin de la Société Préhistorique de France. Tome 30, N. 10: 543-559. RICHTER, D., WAIBLINGER, J., RINK, W.J. & WAGNER, G.A. 2000: Thermoluminescence, Electron Spin Resonance and 14C-dating of the Late Middle and Early Upper Palaeolithic Site of Geißenklösterle Cave in Southern Germany. Journal of Archaeological Science 27: 71–89. RIEK, G. 1933: Les civilisations paléolithiques du Vogelherd près de Stetten-ob-Lonetal (Würtemberg). Préhistoire. Librairie Ernest Leroux. Paris. Tome II. Fascicule II: 149-181. 19 RIEK, G. 1973: Das Paläolithikum der Brillenhöhle bei Blaubeuren (Schwäbische Alb), Teil I. Stuttgart: Verlag Müller & Gräff. SCHEER, A. 1999: The Gravettian in Southwest Germany: stylistic features, raw material resources and settlement patterns. In Hunters of the Golden Age”. Analecta Praehistorica Leidensia 31. Edited by Wil Roebroeks, Margherita Mussi, Jiri Svodoba & Kelly Fennema: 257-270. SCHMID, E., mit Beiträgen von HAHN, J., WOLF & U. 1989: Die altsteinzeitliche Elfenbeinstatuette aus der Höhle Stadel im Hohlenstein bei Asselfingen, Alb-Donau-Kreis. E. Schweizerbart’sche Verlag (Nägele und Obermiller). DE SONNEVILLE-BORDES, D. (1965) :. “Observations statistiques sur l'Aurignacien du Vogelherd, Lonetal (Wurtemberg). Fouilles G. Riek. Fundberichte aus Schwaben, N.F. 17, Mélanges Riek: 69-75. SPANEL, T. 2014: Frauenfigur - üppig oder abstract. Südwestpresse 27/9/2014. SÜDWEST PRESSE. 2010: Abstraktes aus dem Lonetal: Die Venus vom Vogelherd. Südwest Presse. 27.03.2010. Art 4308,420981. WEHRBERGER, K. (ed.) 2013: The Return of the Lion Man. History, Myth, Magic. Thorbecke Verlag. Ulm Museum. WENIGER, G.C. 1989: The Magdalenian in Western Central Europe: Settlement Pattern and Regionality. Journal of World Prehistory, Vol. 3, No. 3: 323-372. 20