New Data from Willendorf and the Beginning of the classic Aurignacian in central Europe



aurignacian aggsbach lamelleaurignacian lamelle scraper

This is a classic Aurignacian core for the detachment of bladelets. The question of the beginning of this technocomplex and its origin is intensively discussed.  It can be assumed with sufficient certainty that the Aurignacian and the Protoaurignacian in Europe is closely related to the dispersal of AMHs. New data suggest a rapid dispersal of both technocomplexes across Europe at c. 44-42 k.a. cal BP.

The origin of these new blade/bladelet technologies remains completely open. The Protoaurignacian has some affinities to the Ahmarian in the Levant (which seems not to be systematically older than the Protoaurignacian!) and part of a common early Upper Paleolithic tradition along the Mediterranean. The classic Aurignacian has no forerunners elsewhere and might be well the invention of a local European network, with oldest data in central Europe. We should not necessarily think on migrations to describe this phenomenon.

The long awaited C-14 dates for the oldest Aurignacian at Willendorf II (stratum 3) are now available and confirm older dates from the 1980ies, which were questioned suggesting a contamination of the samples by older material from Willendorf II; stratum 2 (maybe Szeletian?).  By using stratigraphic, palaeoenvironmental, and chronological data, AH 3 is securely ascribed to the onset of Greenland Interstadial 11, around 43,5 k.a. cal B.P., and thus is older than any other Aurignacian assemblage in the world.

The determination of calibrated C-14 data has become much more reliable during the last years. This begins with intelligent sampling of material from clearly defined archaeological and geochronical defined strata, at the best on humanly modified materials. In addition, new materials, not datable before, for example deliberately perforated marine shells, which are markers of early symbolic behavior can also be dated and the results of such measurements have entered just the field. New material-specific techniques were designed and developed, such as the ultra-filtration of bone collagen, the cleaning of charcoal with ABOx-SC, and the dating of compound-specific bio markers that promise contaminant-free dates. Finally improved statistical tools, such as Bayesian analysis, used in the modelling of the results and several calibration curves for correcting radiocarbon determinations older than 26 k.a cal BP allow more reliable results for Pleistocene-aged material. It has to be mentioned, that a Bayesian approach needs a strictly stratigraphic control of the items that are to be dated.

Distal tephras can provide archaeologists and geoscientists with valuable chronostratigraphic markers.  These markers become especially important when they are found in relation to Paleolithic material that lies close to or beyond the range of radiocarbon dating. In this respect the CI Y-5 tephra in South East Europe, dated to than 39,3 k.a. cal BP is of overall importance.

In S-Europe a Protoaurignacian or a Uluzzian always lies below an Aurignacian, if both entities are present. Wherever the (CI) Y-5 tephra marker is present (South East Europe), the classic Aurignacian consistently overlies the Campanian Ignimbrite. In these parts of Europe the classic Aurignacian therefore seems to be relative young. The Mediterranean Protoaurignacian (at Castelcivita) and Uluzzian (at Castelcivita and Cavallo) and the “Transitional” Paleolithic industries of the Kostenki area (loci 14 and 17) are found below the tephra and must therefore be older than 39,3 k.a. cal BP. These data are affirmed by the fact that pretreated-AMS dated-C-14 samples at of the Protoaurignacian in Italy situate this techno complex at ca. 42-40 k.a. cal BP (for example Mochi G). One of the very few Italian Protoaurignacian contexts which has been directly and precisely dated using modern methodologies is layer A2 of Fumane Cave. Bone and charcoal samples from A2 date to ca 35,5 k.a. BP and the start of this earliest Aurignacian phase is calculated to fall between 41.8 and 40.8 k.a. cal BP.  According to the sparse data available, the Uluzzian dates to  43 to 41 k.a. cal BP.

In S/W- France the earliest Aurignacian of Abri Pataud dates slightly later to around 41-40 k.a cal BP.  Other sites in the Aquitaine seem to be younger. Towards the Paris basin, at Les Cottes the Protoaurignacian is dated to a short episode around 39 and early Aurignacian around 39-36 k.a.

Central Europe: The heavily debated lowermost Aurignacian levels at Geissenklösterle (AHIII) in the Swabian Jura dates to 42,9- 39,9 cal B.P k.a cal BP if we take for sure that AHIII is an archaeological reality and not a secondary reconstruction bias. The chronostratigraphic position of AH 3 (Willendorf II) is now the best evidence for a an early Aurignacian technology in Central Europe at least slightly before 43,3 k.a. cal B.P. Maybe the nearby Krems site was older and contained both Protoaurignacian and Aurignacian-but this information is lost forever ( . Other Aurignacian sites in Austria and Germany are of later age (except possibly Keilberg-Kirche and Senftenberg). An early date in this context is the Aurignacian bone point of Pesko in Hungary, dated  to between 41,7 and 40,2 k.a. cal B.P. It remains exciting- stay tuned!

 Suggested Reading:


Posted in Plaeolithics and Neolithics | Tagged , , , , , , , | Leave a comment

The early Gravettian in Central France: La Vigne Brun

aggsbachs villerest2

This is a 3, 5 cm (Micro) Gravette point from the Vigne Brun site, found at the site in large quantities (although mostly broken), and called: “Microgravette of Vigne-Brun type” ( The old C-14 dates from La Vigne Brun fall around 23 k.a. BP and are not taken as valid. Technologically and typologically the site shows a pattern that is characteristic for the early Gravettian in France and Southern Central Europe. Calibrated C-14 dates will probably soon resolve the chronological confusion about this key locality.

La Vigne-Brun is located in the eastern Massif Central, 5 km upstream from Roanne in the Loire river valley. It is one of several open-air Palaeolithic sites at a locality called Saut-du-Perron (Villerest district). Other important sites nearby are Le Roche de la Caille, la Goutte-Roffat (a Magdalenian with 182 engraved schist plaquettes) and the Champ- Grand site (Quina Mousterian). These sites have been explored for more than a hundred years and lamentably were flooded in 1983 following the construction of the Villerest dam. From 1977 to 1983 an area of 470 m² at La Vigne-Brun was investigated by J. Combier and, despite being the object of a salvage excavation, the site was fully excavated under relatively good conditions.

aggsbachVigne Brun is known in the archaeological literature principally for the discovery of a series of habitation structures associated with a Gravettian industry. During the last years meticulous typological, technological and economic approaches were able to reconstruct the systems of lithic artefact manufacture at the site.  These artifacts comprise dihedral burins, endscrapers and a large lot of projectiles (Font Robert points, Gravettes and Microgravettes). The bladelet blanks used for the Microgravettes were obtained using three chaînes opératoires: one exclusively for bladelets, one incorporating small blades and bladelets and a third, uninterrupted sequence from large blades to small blades to bladelets. Raw materials are largely dominated by a regionally available “jasper” flint. The predominant exogenous flint is a yellowish-grey to brown variety of excellent quality which was obtained from lower Turonian outcrops in the Gien region. The procurement pattern of raw materials observed for unit KL19 at La Vigne-Brun does not particularly resemble the “classic” pattern observed at many Upper Palaeolithic sites, where there is a contrast between only slightly modified local materials and others transported as finished products from a distance. At La Vigne-Brun procurement strategies are equally variable for both local material and that brought from further away. Regional flint was imported to the site from a distance of 5 to 30 km either as blocks or as burins and Gravettes/ Microgravettes followed by splintered pieces and endscrapers.

The early Gravettian of Vigne Brun shows both technological and typological correspondence to other sites in France and the Swabian Jura: The open air site of Le Sire (Mirefleurs, Dpt. Puy-de-Dôme), was attributed to the early Gravettian, based on one uncalibrated date of 27,300 +/- 330 BP. Beside an important proportion of Microgravettes (36% of the tools), some Font-Robert Points and Flechettes are also present. The lower Gravettian lithic ensemble of the open-air site Azé-Camping de Rizerolles(Saône-et-Loire, France)  is composed of over 48.000 pieces, among them Gravette points, Microgravettes, Nanogravettes and the fragment of two Font- Robert points. In the Swabian Ach valley (Geißenklösterle and Hohle Fels) we also find an early Gravettian (29-27 k.a. BP) focused on the production of Microgravettes with some Flechettes and isolated Font-Robert points.

gravette aggsbach

Suggested Readings:


Posted in Plaeolithics and Neolithics | Tagged , , , , | 1 Comment

Levallois- Mousterian in S-Italy


italia mousterian aggsbach

On the left this picture shows two Quina scrapers from the Gargano and from Venosa (Basilicata). While the Gargano has been introduced during earlier posts ( – the Basilicata also known as Lucania is a region in the south of Italy, bordering on Campania to the west, Apulia (Puglia) to the north and east, and Calabria to the south/west. The artifacts on the right are typical Levallois-based artifacts from a surface site in the Gargano. Two of them are pretty blade based and have affinity to the late Mousterian in N-Italy immediately before the arrival of AMH.

New excavations in N and S-Italy have shown, that the evolution of the Italian Middle Paleolithic follows in large lines the evolution of this complex in S-France. At the Grotta di Fumane, an important site in northern Italy, which has been extensively explored over the last two decades a succession of Mousterian- Uluzzian-Protoaurignacian and Aurignacian (and early Aurignacin parietal paintings  and an ochered shell in the late Mousterian !) was documented and extensively dated ( The 12 m thick sedimentary sequence are related to repeat and complex human occupations. The earliest assemblages record the almost exclusive use of the Levallois method (MIS 5-4) for the production of flakes with unidirectional and centripetal recurrent modalities. The first striking technological replacement occurs in BR6, up to BR3, where there is a complex of layers with the infrequent occurrence of bones, flakes and scrapers, made with a method closely resembling the Quina technique (MIS4). Further evidence of variability in lithic technology is provided by the re-appearance of the Levallois technology during MIS3, in layer A11 – although here focused more on blades than flakes – and by the Levallois/Discoid alternance throughout the transect from A10V to A5-A6 (MIS3).

Guado San Nicola (Monteroduni, Molise) was recently dated to the MIS 11-10 boundary (see comments) by the 40Ar/39Ar method and provided an ensemble of Levallois products within an Acheulean context. This is older than the recently reported “earliest Levalloisian from Eurasia” at Nor Geghi 1, Armenia (OIS10/9e boundary). Nor Geghi 1, Armenia, was recently reported being the earliest site with synchronic use of bifacial and Levallois technology outside Africa. Units 5 to 4 at this site are correlated with late OIS 10/early OIS 9e, whereas Units 3 to 2 with OIS 9e (335 to 325 k.a.)

In North Italy the oldest Levalloisian was reported from  San Bernardino Cave (MIS8/7 boundary).

Although the Levallois method was suggested to be absent in S-Italia during MIS5-3, stratified sites with Levallois based ensembles and absolute dates have been excavated during the last two decades, contradicting the view, that the Italian South did not share the technical and cultural innovations in the rest of Europe. The Levallois technique during the last interglacial/glacial cycle (MIS5 and later) is now well established, especially in Puglia.  For example, the layer B of the Grotta Bernardini which recently was dated to a period just after 108 k.a. shows basically a Levallois mode of lithic production, as the industry  of  the G layer  at Grotta Romanelli  (MIS 4/3) and corresponding strata of the  Grotta Titti.

Suggested Readings:

A closer view on one of the Gargano retouched Levallois blades – a stiking convergence to the ca. 220 k.a. old Hummalian artifacts in the Syrian Desert! (


Posted in Plaeolithics and Neolithics | Tagged , , , , , | 2 Comments

100-year Anniversary: Peyrony at the lower Rock shelter of Le Moustier

aggsbachs moustier

This is a classic Mousterian point, made on a thick Levallois point (8,5 cm long)  from the MTA-levels (Layer G) of the type site.

Le Moustier is an archeological site consisting of two rock shelters in Peyzac-le-Moustier, belonging to the community of Saint-Léon-sur-Vézère; Dordogne, France. Le Moustier lies about 10 kilometers northeast of Les Eyzies along the Vézère River.

The upper shelter was first excavated from 1863 by Henry Christy and Édouard Lartet, while the lower shelter was first opened for excavations by Otto Hauser during 1907. He discovered the famous Le Moustier Neanderthal adolescent in 1908. He sold it to the Völkerkundemuseum in Berlin, where most of the postcranial skeleton was destroyed during the bombing raids during the late WW II. The skull and mandible survived and the cranio- dental skeleton of Le Moustier 1 represents perhaps the most complete Late Pleistocene adolescent hominid skull. A second skeleton was later found at the site by Peyrony: Le Moustier 2, a Neanderthal neonate.

Peyrony excavated the lower abri at Le Moustier during 1913-1914 along the western edge of the earlier Hauser explorations. He established an important sequence of Paleolithic strata traditionally numbered from A to L. A: sterile layer, B: Typical Mousterian; C/D. quasi sterile; E: sterile layer;  G: MTA-A (around 50-55 k.a BP*); H: MTA-B (around 42-48 k.a BP*); I: denticulated Mousterian (around 40,9 k.a BP*); J:, Typical Mousterian (around 40,3 k.a BP*);  K: cryoturbated and mixed Castelperronian / Mousterian (around 42,6 k.a BP*); L: Aurignacian ; (*Dates by TL). The most recent excavations were conducted by Laville and Rigaud in 1969.

The chaine operatoire of the MTA assemblage from Le Moustier  is characterized by a recurrent centripetal unidirectional Levallois concept with production of a series of unidirectional blanks. The Le Moustier G-ensemble is mainly composed of Senonian flint which displays gray to black colors. This flint is also known to have been used by Neanderthals at several nearby sites (La Rochette, Le Ruth, surface collection at Plazac…) and widely during the upper Paleolithic of the Vezere valley.  Cordiform flat handaxes (4-14 cm long) were found in abundance together with large quantities of simple side scrapers. Quina scrapers and convergent tools were rare.  Indeed, Le Moustier G together with the nearby La Rochette are considered as biface workshops, where tools were manufactured in advance to be brought eventually to other sites.

Levallois blanks from the MTA at Le Moustier G often have a laminar aspect, but these blanks are broader that Upper Paleolithic blades as you can see in the picture below .  It seems that they often were used without any retouches and for the production of backed pieces, which often are quadrangular in shape or had a curved back (Abri Audi knifes).

A large number of retouched Levallois points are known from Le Moustier G. Many of them have a bilateral ventral retouche, often of a unilateral quite steep aspect. This led Peyrony to suggest, that many of these “Points” were in reality rather knifes than projectiles.

The second picture shows two handaxes and a typical “knife”, along with the elongated point of the first picture from the Le Moustier site-all made from the raw material mentioned above. It is a pity, that 100 years after the excavation of Peyrony we have no comprehensive monograph of the site. Peyrony s report from 1930 is of poor quality (PEYRONY, D. (1930). Le Moustier. Ses gisements,  ses industrlos, ses couches géologiques, in  Revue Anthropologique, vol. 40, pp. 48-76).


le moustier aggsbach


Suggested Reading:

Look at the rich material you can get by the steadily growing and updated Paleolithic site in the net:

Don was a great help for me when I started this blog in 2010…

Le Moustier during the early 20th century:

hauser le moustier


Posted in Plaeolithics and Neolithics | Tagged , , , , , | Leave a comment

The Gravettian at the Solutré Site


This is a highly patinated fragment of a “Spitzklinge“ (pointed blade) from the Gravettian at the famous Solutré  site- almost identical to examples from the Willendorf site in the Wachau region. Pointed blades have been identified as a major component of the  “Facies II” of the Pavlovian in Central Europe, which is characterized by the use of marginal retouches, pointed blades and by a reduced number of microliths (Předmostí; Dolní Věstonice II middle and upper strata, Willendorf II; 6/7/8 and Langenlois). The second picture shows two other fragments of retouched and pointed blades and a burin from the Gravettian of the site:

solutre gravettian

The Gravettian of the central-eastern region of France, geographically bridging central and South Europe, is concentrated in the valleys of the River Loire and of the River Saône, in the south of Burgundy. Lithic series were principally collected on the surface, and only three major have contextual informations about the Gravettian occupations of this region: the site of La Vigne-Brun, in the Loire valley and the sites of Azé and Solutré along the Saône valley.

Solutre, a well-known kill site, occupies a vast area at the foot of an escarpment. The slope deposits contained a series of archaeological layers representing most of the Paleolithic cultures from Mousterian to Magdalenian. The massive accumulation of horse bones under the escarpment, unique in the European archaeological record, intrigued both archaeologists and the general public. Inspired by reports of bison kills in North America, Adrien Arcelin proposed in a novel, an interpretation which attracted a great deal of attention. In this popular narrative, Paleolithic hunters chased horse herds up to the top of the escarpment and forced them to jump to their death (

Currently accepted view proposes that hunters intercepted animal herds as they moved through the Solutré valley during their seasonal transhumance from the Alluvial Plain of the Saône to the Macônnais Uplands.  They forced their prey into natural rock traps along the southern flank of the Roche just under the falt line where they could be slaughtered. 

The lithic material at Solutré is highly patinated (96% of the parts). Especially during the Gravettian, postdepisitional disturbances of the Archeological material by trampling and cryoturbation, is present. The limitation of old biased collections makes it difficult to reconstruct the operational sequences. 

From a typological view, J. Combier attributed the Gravettian industry at Solutré to a final phase of the Gravettian. The low number of tools (old and new excavations) should lead to some caution. Moreover, the entire cluster including Microgravettes, a point of La Gravette, a La Font-Robert along with some endscrapers and burins could be more reasonable linked to an early Gravettian. The view of an early Gravettian is also consistent with the MSA data (Sector J10: 28,420 ± 160 BP; Sector L13: 28 280 ± 150 [Montet-White et al. 2002 p.186]).

Suggested Reading:


Posted in Plaeolithics and Neolithics | Tagged , , , , | 1 Comment

Quina Mousterian from the Gargano


quinaquinaThis is a Quina scraper made of high quality flint, found near Mattinata in the Gargano together with a pure Quina Mousterian ensemble. The scraper combines a scraping edge, covered with a „Retouche écailleuses scalariforme“ and an a oposite edge, allowing a handhold use or the hafting of the artifact. Intentional backing of stone tools in Europe was known since the Acheulian (

quina aggsbach2Neanderthals successfully colonized the Italian peninsula, coping with different ecotones / niches and climates during a considerable time since the Middle Pleistocene. As already known from Northern France, Middle Paleolithic industries and regional differentiation of the lithic industries in Italy appeared long before the last interglacial ( They are assigned to a late Acheulian with or without a Levallois component rare in bifaces (Pagicci in Apulia, Torre in Pietra-Level D [Rome], Levallois- Acheulian of the N/E-loess belt;

The Yabroudian of the Near East is a flake industry with minor production of blades dominated by the conspicuous presence of a Quina Chaîne Opératoire for the shaping of scrapers. These scrapers are quite distinctive and well known from Middle Paleolithic Mousterian of Europe, however the Yabroudian is much older than the European manifestations and shows the Quina phenomenon and scrapers as early as (420-200 k.a.) in the Levant. It is thus quite untenable to suggest any kind of connection between the two very similar or almost identical, EuropeanQuina phenomena (Bakai 2013)..

Quina Mousterian sites in Europe are usually dated to MIS 4 and early MIS 3. Technologically and typologically, Quina ensembles are chacterized by the near total absence of Levallois technology and are dominated by single side-scrapers and transverse scrapers and Limaces with typical Quina type, stepped retouch.  At the type site La Quina (Charente) large transversal scrapers up to 20 cm long with invasive retouches, called by Henri Martin: “hachoirs” are present. Many of them are made on bifacial blanks, although their frequency may be exaggerated by a selection bias from older collections. The so called Bola stones, while not abundant, are well known from the Quina Mousterian and are also present in the Quina levels at Combe-Grenal and Chez-Pinaud Jonzac. Based on the probable traces of impact on their most prominent points, their size and their weight, it seems likely that they may have served as hammer stones. Such items are only known from the French South-West.

aggsbachs scraper backedIn Italy, the Quina technological system  was dominant during the “Pontinian” in the Latium region where the raw material consisted of small flint pebbles, which were not usable for the Levallois technique, but of excellent knapping quality.

For a long time the Levallois technique seemed to prevalent only in the North-East while in Southern Italy the Quina system was suggested to be omnipresent.  It is now clear, that this view was biased (

In contrast to other S-Italian sites, in the Gargano region high quality chert was always available during OIS4/3 . The Levallois technique is attested at Piani di San Vito (an open air site) and the Grotta di Spagnoli (a cave site) and at many non-stratified sites in the region. On the other hand, excellent raw material and large cobbles of flint, do not preclude the use of a pure Quina system, as shown by surface findings from an open air site at the Monte Sacro (north of Mattinata). At this locality a classic Quina ensemble was found during 30 years of meticulous surface collection. The Quina scraper shown here  is one example from this site. 


Posted in Plaeolithics and Neolithics | Tagged , , , , | Leave a comment

A Mousterian Point from Le Placard / Charente


aggsbach katzman placard

This is an convergent Middle Paleolithic artifact (usually called Mousterian point with Quina retouche) from an early excavation at the famous Placard cave in the Charente. Le Placrd could have served as one reference site for the region if it only had beeen properly excavated, which was definite not the case. Anyhow its stratigraphy played an important role in the famous  “bataille aurignacienne” between the priest-scientists around  the Abbe Breuil and the Materialistic Prehistorians, followers of G and A de Mortillet during the early 20th century.

The Grotte du Placard, originally called Grotte à Rochebertier (Commune de Vilhonneur; Charente), is a large cave on the left bank of the Tardoire 13 kilometers upstream from La Rochefoucauld. Carved into a limestone cliff, the entrance to the cave is 17 meters above the river level, with a wide porch giving access to a room 25 meters long and 10 wide at the entrance. It is itself extended to the right by two narrow galleries, and to the left by a more important gallery. There are several smaller grottoes with Paleolithic material spanning  a time span from the Chatelperronian (?) to the Magdalenian in the immediate vicinity (Grotte de Moradie [Magdalenien], de La Combe, de l` Ammonite [Chatelperronien?; Magdalenien final] and du Sureau).

It seems that the first excavations at the Placard cave took place in 1853. The Placard site was first mentioned in the literature by J. Fermond in 1864. He was totally focused on the extraordinary rich middle and upper Solutrian levels of the site. During the later years A. de Maret, the land owner, made extensive and disastrous excavations and virtually emptied most of the cave fillings. He remains the “bete noir” of the excavators at Placard, because his technique of extracting the artifacts was below any standards of his time. He was on hunt for first-quality pieces and destructed all other artifacts to prevent others to sell pieces from Placard and to keep the prices high. It is suggested that thousand shouldered points from the Solutrean were lost during this destructive “curation”.  Later more methodic excavations by Fermond, Masfrand, A. de Mortillet and by Chauvet followed. Finally the deposit, which was without doubt one of the richest in Western Europe, was not subject to any protection and suffered numerous digs that fed many museums and private collections for almost one half of a century. Therefore J. Roche in 1958 made an attempt to reconstruct the stratigraphy from what was left (mainly Middle Paleolithic) after 80 years of devastation. As a result of renewed illegal excavations, Duport resumed the excavations in 1988 and the discovery of rock art panels has led to the opening of a rescue campaign by J. Clottes and others.

A. Masferad described a course stratigraphy in 1902:

  1. Magdalenian (1,5 m thickness)
  2. Solutrean (1 m thickness)
  3. Mousterien (1,5 m thickness)

During his early excavations, Maret he had detected one stratum of Mousterian, two strata of Solutrean (a middle Solutrean with large leaf shaped points followed by a rich upper Solutrean with thousands of shouldered points) and four Magdalenian strata (beginning with a Badegoulien / early Magdalenian a raclettes, topped by a Middle Magdalenian and traces of a “Magdalenien final”).

Based on Marets communication, the “official “and somewhat theoretical stratigraphy was published by A. de Mortillet in 1906. He described a “Robenhausien” on the top and below, three Magdalenian levels, two Solutrean levels and, at the base the Mousterian. On Mortillets stratigraphy, each level was isolated by a sterile layer formed by the debris, which fell from the cave ceiling.

The Mousterian of Placard is testified by a rich collection of artifacts characterized by a non- Levallois production system of the blanks. Thick flakes were modified into convergent tools (convergent scrapers and “points”). There are many transversal scrapers with a typical “Quina retouche”. Roche described in total five strata of a Charentian (Quina and an industry between Quina and Ferrassie), seven strata with a typical Mousterian and one with a denticulated Mousterien. Roche was able to identify 17 Mousterian strata during his final excavations, some of them only 5 cm thick. This observation gives us an idea about the informations that were lost during the early diggings….

A small Middle Solutrean point from the site. A gift from A de Mortillet to the Austrian Prehistorian Hoernes who gave it to my Grandfather around 1905:



Suggested webpage:

Posted in Plaeolithics and Neolithics | Tagged , , , , , , , | Leave a comment