The Acheulian site of Cantera Vieja (Madrid, Spain) and the Lower to Middle Palaeolithic transition in central Spain

doi:10.21203/rs.3.rs-4195503/v1

Download PDF

Research Article

The Acheulian site of Cantera Vieja (Madrid, Spain) and the Lower to Middle Palaeolithic transition in central Spain

https://doi.org/10.21203/rs.3.rs-4195503/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

The Cantera Vieja archaeological site is situated in Madrid, Spain, an area rich in Miocene flint and that has revealed several quarries at Lower and Middle Palaeolithic sites in recent years. Our study presents a multidisciplinary examination of the site, featuring an analysis of its geological context, site formation processes, chronological framework, and assemblage characteristics. Formed on the interfluvial platform between the Manzanares and Jarama Rivers, the Cantera Vieja site was created during the late Middle Pleistocene (194 ± 13 to 215 ± 16 ka) by alternating alluvial and mass-movement colluvial processes. The site boasts a typical Acheulian assemblage, characterized by numerous handaxes and preforms, and bifacial shaping flakes and fragments, with a limited number of flake supports, non-Levallois flaking elements, and a few final retouched tools. However, comparison of the archaeological assemblages across the Iberian Peninsula, including Cantera Vieja, with those reported from other contemporaneous sites across different parts of Europe indicates that this technocomplex could have slight location-specific differences in its industrial evolution. Our study at Cantera Vieja allows us to discuss the presence of specific lithic expression in the context of the western Acheulian in general and its transition to the Middle Palaeolithic in Iberia in particular. At the same time, the characteristics of the Cantera Vieja assemblage provide potentially generalizable insights into the social organization of lithic production during the late Middle Pleistocene. We consider that Cantera Vieja must have fulfilled an essential role as a place for training and learning biface knapping.

Pleistocene

Acheulian

Mousterian

Palaeolithic

biface

quarries

skills

Understanding how lithic technocomplexes change over time is crucial for understanding the cultural evolution of different human species. The Acheulian first appeared in Africa around 1.95 million years ago. It has been found at sites such as Garba IV (Mussi et al., 2023), Kokiselei (Lepre et al., 2011), Konso KGA6 (Beyene et al., 2013), Gona (Stout et al., 2010) and FLK West (Díez-Martín et al., 2015), and in some of these examples the Acheulian tradition is combined with different technological expressions (de la Torre, 2016; Mussi et al., 2023). The ages of these African sites find echoes in the last attestations found on the Iberian Peninsula close to Marine Isotope Stage (MIS) 5–6, which, despite their much more recent chronologies, still display clear typological influences from Africa and also from much earlier sites in central Europe (Santonja and Villa, 2006; Santonja and Pérez-González, 2010; Méndez-Quintas et al., 2018, 2019).

The evolution of the Acheulian, which lasted at least one and a half million years, is likely to have undergone significant changes and variations, probably suffering reversals and veering off down detours at times (Santonja and Villa, 2006; Sharon et al., 2011). Exploring these changes, or the stasis occasioned by the lack of changes, as the case may be, can be achieved by comparing assemblages. While the development of lithic exploitation strategies is an essential aspect of technocomplex evolution (Herisson et al., 2016; Rubio-Jara et al., 2016; Sharon and Sharon, 2016), this study focuses on a particular example of the Acheulian tradition from central Iberia. This geographical context is home to an important chronological enduring of the bifacial tradition, as confirmed by western European sites that represent the starting point of the Acheulian presence in Europe (Scott and Gibert, 2009; Walter and Trauth, 2011; Jiménez-Arenas et al., 2011; Mosquera et al., 2015, 2016; Bourguignon et al., 2016; Moncel and Schreve, 2016; Moncel et al., 2019, 2021; Lewis et al., 2021; García-Medrano et al., 2023; Ollé et al., 2023), as well as witnessing the persistence of the last vestiges of this tradition (Méndez-Quintas et al., 2019, 2020).

The central region of the Iberian Peninsula is an interesting area for studying Palaeolithic cultural evolution. The ecological context of the area, dominated by a complex structure of terraces associated with the Tagus River tributaries, has played host to some of the most significant Lower and Middle Palaeolithic sites in Spain (Rubio and Panera, 2002). Historically, this region has yielded several sites related to the use of macro mammal carcasses in riverine contexts, and some of the latest known uses of elephants (Panera et al., 2014; Rubio-Jara et al., 2016; Santonja et al., 2016). More recently, several sites discovered in the interfluvial areas of the Manzanares and Jarama Rivers have revealed interesting associations between human tool production and the flint-quarrying contex (Baena et al., 2015; Bárez et al., 2016; Baena Preysler et al., 2018).

Cantera Vieja is one of the most relevant examples of Acheulian sites recently excavated in the central Iberia region. This site provides a significant assemblage, mainly related to the exhaustive production of bifaces. It documents the concurrence of different knapping skill levels and offers examples of several bifacial stages. The current study examines the geoarchaeological context of the site, its formation processes, the chronological framework, and the lithic productive behaviour found there. The aim of this study is to improve knowledge of the Acheulian tradition and to outline its implications in the regional context.

2.1 Dating methodology

Three luminescence dating samples were collected from the sedimentary sequence at Cantera Vieja. Two were collected from above and below archaeological level 1 – sample CVJ18-3 (above level 1) and sample CVJ18-1 (underlying level 1) – and one sample, CVJ18-2, was collected from the bottom of archaeological level 2. The three sediment samples were dated using multi-grain K-feldspar post-infrared infrared-stimulated luminescence (pIR-IR) (Buylaert et al., 2012). Samples CVJ18-3 and CVJ18-1 were additionally dated using single-grain quartz thermally transferred optically stimulated luminescence (TT-OSL) (Wang and Wintle, 2006; Arnold et al., 2015). Details about sample preparation, instrumentation, measurement protocols and data analysis are provided in the Supplementary Information, and follow the approaches used in previous studies that have applied these two methods in tandem (Demuro et al., 2014, 2015; Terradillos-Bernal et al., 2023). The K-feldspar pIR-IR D_e measurements were made with 360-grain aliquots using a stimulation temperature of 225°C (herein denoted pIR-IR₂₂₅), following the outcomes of suitability assessments performed on sample CVJ18-1, as described in the Supplementary Information. The TT-OSL signals were measured for individual quartz grains following the procedures detailed in Demuro et al. (2014, 2015), which yielded suitable laboratory validation test results for sample CVJ18-1 (see Supplementary Information tables from S1 to S9 and figures S1 to S3 ).

Dose rate evaluations were made using a combination of in situ gamma spectrometry measurements (Arnold et al., 2012; Duval and Arnold, 2013), low-level beta counting and use of a long-term water content that equated to 13–15%, as well as assessments of internal beta dose rates and cosmic ray contributions. Additionally, high-resolution gamma spectrometry (HRGS) measurements were made on the three sediment samples to assess the state of secular equilibrium in their ²³⁸U and ²³²Th decay series. Full details of the dose rate estimation procedures are given in the Supplementary Information.

2.2 Geoarchaeological and micromorphological study

To complete the geomorphological and stratigraphic study, data were compiled from previous publications, detailed topographical plans, geological maps, and aerial photographs of the area (from the 1956 American flight). Field stratigraphic data were also obtained to characterize the different levels documented here. Two samples were taken for micromorphological study, complementing the detailed stratigraphic study to document the sediments’ diagenetic processes and environmental origin. The samples were collected from the southern profile of the site (see Supplementary Information figures S4 to S7 and table S10 ) after covering them with plaster. The first block was named CV1 and was 25 cm long, incorporating stratigraphic units I.3 and II.1. The second block, defined as CV2, was 21 cm long and incorporated archaeological level 1 (unit V.3), level R (units V.1 and V.2) and part of level 2B (units IV.3 and IV.4). Samples were processed in the micromorphological lab of the Centro Nacional de Investigación sobre la Evolución Humana (CENIEH), Spain.

2.3 Archaeological fieldwork and lithic analysis methodology

The archaeological fieldwork was undertaken in 2018 under the supervision of the urbanization project of the Los Berrocales district, projects known as PAUs (Polígono de Actuación Urbanística), and consisting of considerable areas destined for increased urban development in the surrounds of Madrid city (Baena et al., 2008, 2011; Bárez et al., 2016). When surveying such large areas, the archaeological methodology to detect potential Palaeolithic sites necessitates the creation of a network of trenches that are sunk through the Quaternary deposits. The Cantera Vieja fieldwork started with the supervision of those trenches as they were dug, and continued with the documentation of abundant fresh lithic industries emerging from the new profiles in a narrow area near the A-3 Highway.

An initial survey was conducted, which resulted in an approximate total surface area of 48 m². However, due to the urgent nature of the excavation, we were only able to excavate a sub-area of 30 m², which included all the archaeological levels down to the underlying Neogene sediments. The remaining area of 18 m² was excavated in the upper levels only, and the archaeological environment was declared a protected area (see Fig. 1).

The archaeological excavation resulted in a total of 6352 coordinated lithic pieces (with x, y and z position). Due to time constraints, small lithic remains less than 3 cm long were left in reference sectors of 50 × 50 cm, along with some larger pieces. The small remains, which included flakes, fragments and residues, totalled 26,663 pieces, out of which 16,738 belonged to level 1, 9679 to levels 2A and 2B, and 246 to level R. However, the total number of pieces was 33,015, since the coordinated lithic industry (n = 6352) have to be included in the total sum.

Spatial analysis was carried out by the integration of all coordinated pieces as points into a GIS (ArcGIS 10.8.2 and later ArcGIS Pro) and by representing and analysing a combined selection of archaeological level, lithic category, morphological dimensions, and also on-ground orientation and slope of pieces.

This lithic assemblage study is based on identifying lithic categories from different configurations and flaking concepts and methods. This identification is based on a comparative analysis of archaeological and experimental replicative collections (Baena and Cuartero, 2006) produced at the experimental laboratory (Laboratorio de Arqueología Experimental LAEX) of the Universidad Autónoma of Madrid (UAM). Its application identifies techniques of production, methods employed, and the presence of knapping-skill technotypes (Torres Navas, 2022). These knapping skill levels are one of the most discussed aspects in comparative studies (Delagnes et al., 2023). Our definition of each one (novice, apprentice, advanced apprentice, and expert) is based on three different aspects: technical skill, technological knowledge, and the ability to put the knapping process into practice (Tixier et al., 1980). We maintain strict control of the level during experimentation, to the point that some agents can even be moved from one category to the other after several experiments due to their acquisition of experience.

At Cantera Vieja, the presence of a large number of handaxes and early-stage biface production pieces has led to the question of whether this site was used for a specific purpose involving use of these tools or simply for the production of these types of artefacts. To better understand the meaning of these lithic sets, several lines of enquiry have been pursued, leading to a conclusion about the site.

Functional studies have been conducted, including the creation of an experimental reference collection using flint from the site. By analysing the patterns of use wear resulting from various experiments, we were also able to examine and distinguish between the traces resulting from post-sedimentary effects and those arising from the use of tools. The result was a collection of images of different traces of use that could be compared via microscopic observation with the archaeological material. In total, 216 archaeological lithic tools were analysed in the Cantera Vieja collection: 67 handaxes, 57 preforms, 44 simple flakes, 40 retouched flakes, 7 cores and 1 cleaver.

Our study of the Cantera Vieja lithic assemblage was challenging due to the presence of carbonates and silicates. To overcome this, various cleaning methods were developed using different types of acids. The most effective method involved immersing the archaeological piece in deionized water and then using an ultrasonic bath for a sufficient period of time. This process allows the carbonate crusts to fall away from the piece.

After this, it was possible to proceed with the observation of the material, which was undertaken by using an Olympus DSX1000 digital microscope for macroscopic observation of the knapping marks and a Leica DMRX microscope for microscopic analysis of the use wearAdditionally, we conducted a morphometric study by scanning all the Cantera Vieja preforms and other similar tools from the San Isidro collection (with permission of Spain’s Museo Arqueológico Nacional) at high resolution using a NextEngine scanner. We then post-processed the files with Scanstudio, MeshLab 2020 v07 and CloudCompare v.2.12. At the same time, we initiated an experimental protocol to produce biface replicas under varying levels of technological expertise at the UAM’s LAEX (see Supplementary Information figure S8). To this end, we recruited several volunteers with varied knapping skills to produce bifaces on the basis of the technical skill, technological know-how, and practical implementation’. We recorded the entire sequence by video and created 3D morphologies of the final pieces. We recruited 10 novices, 10 apprentices, 5 advanced apprentices, and 3 experts, resulting in around 100 experimental models (Torres Navas, 2022).

The Cantera Vieja archaeological site is located in the northwestern area of the Southern Subplateau of the Iberian Peninsula, within the Madrid Basin, one of the geological sub-basins that make up the Cenozoic Tagus Basin. This intraplate continental basin was formed during the Alpine orogeny (Vegas and Banda, 1982), whose deformation reactivated three lineations of mountain ranges that eventually surrounded the basin: namely, the Central System, the Toledo Mountains and the Iberica Chain (Fig. 2).

The deposits that fill the Madrid Basin were produced during the Miocene, by means of a regional system of coalescing alluvial fans, generated in the border zones at the foot of the mountain ranges, under an endorheic or semi-endorheic regime (Calvo et al., 1996; De Vicente et al., 1996). Three tectonic phases also conditioned the formation of three large Miocene tectono-sedimentary units (Alberdi et al., 1983; Junco and Calvo, 1983; Hoyos et al., 1985): namely, the Lower, Intermediate and Upper Miocene units. In each of these large Miocene units, the detrital facies (proximal and distal), the transitional or intermediate facies (lutitic plain and lacustrine margin) and the central or lacustrine facies (evaporitic or carbonate, depending on aridity) were deposited concentrically.

As a consequence of the Alpine orogeny, in addition to the reverse faults on the basin rim, a system of faults was also generated at the bottom of the plinth, under the Miocene basin-fill deposits, which fractured these Miocene materials. The orientation of these faults is reflected on the surface, conditioning during the Quaternary the alignment of most of the fluvial courses, the evolution of their valleys and their sedimentary patterns.

The environment of the Madrid Basin is characterized by two morphological elements: the valleys and the interfluvial areas that separate different regional rivers (Royo Gómez et al., 1929). These dividing surfaces are made up of patches of feldspathic sand ramps. These erosive surfaces are currently found mainly on the culminating arkosic sands of the Miocene Intermediate Unit (MIU). Towards the southeast, these sands pass laterally into clay facies with carbonate and flint layers, which provides a strong lithological-structural control for the southeastern-most area of the basin. As a consequence of the continuous incision of the Quaternary drainage network, these surfaces were reduced, and their remnants are now usually arranged parallel to the direction of the main river valleys.

The geomorphological evolution during the Quaternary of the Manzanares-Jarama Interfluvial Platform (hereinafter MJIP), as well as that of the Manzanares and Jarama River valleys, depends on several factors: among others, climatic variations, changes in the fluvial base level associated with sea level variations due to climatic causes, neotectonics due to isostatic uplift and block adjustments and lithological-structural controls (Pérez-González, 1980, 1994; Silva et al., 1988).

The erosive-structural origin or lithological-structural control of this platform has already been described previously in a range of studies (Silva, 1988; Goy et al., 1989; Silva et al., 1999; Bárez and Pérez-González, 2007; Baena et al., 2008, 2011, 2015). However, it was not until the work of Bárez et al. (2016) that the MJIP was accorded recognition as a geomorphological entity (Figs. 3 and 4).

It is important to consider that before the Quaternary drainage network was embedded, the MJIP formed part of the northeastern end of a larger structural block. After the lower course of the Manzanares River was embedded, this larger structural block was divided into the MJIP at the northeast end and the Espartinas Platform at the southwest with a length four times longer than the MJIP. The geomorphological context of the study area is summarized in Figs. 2, 3 and 4.

Towards the southeast of the MJIP, flint levels are more frequent, thicker and have greater lateral contiguity. Due to the outcrops of the flint/opal levels that characterize the tops and slopes of the hills found on the MJIP, this area has been a recurrent point of capture for raw material (flint) from the Palaeolithic to the Neolithic. This interest in the area’s geology extends to modern mining activity.

According to previous studies (Manzano et al., 2011; López-Recio et al., 2014; Manzano Espinosa et al., 2018), there seems to be a relationship between the geomorphological position and the ages of the existing Palaeolithic sites on the MJIP. Most of the Mousterian sites (Middle Palaeolithic) are located in the high and intermediate zones of the MJIP such as El Cañaveral (Baena et al., 2011, 2015), while the Acheulian sites (Lower Palaeolithic) are located in areas close to the valley floor or in depressions such as Los Ahijones (Figs. 1 and 2). The Cantera Vieja site is located at an average elevation of 640 m, while the Charco Hondo deposit in Los Ahijones is located at 625–630 m altitude (Bárez et al., 2011, 2016) and the El Cañaveral Palaeolithic sites are found at higher elevations (647–650 m).

The Cantera Vieja site is located in the central area of the MJIP, about 2 km southeast of the prominence of Cerro Almodóvar (726 m) and about 2.5 km west of the Los Migueles stream (Fig. 4). The MJIP’s original platform is quite degraded due to the process of uphill erosion by a small tributary of the Los Migueles stream. The hills of Alto del Espinillo or Las Peñuelas (661 m) were also generated as a result of this erosion, and Cantera Vieja is located at the foot of this hill (between 640 and 641 m above sea level), as well as at the base of Cerro del Murmullo (656 m) 500 m to the south, and the Alto del Retiro (667 m) 1 km to the west. On the top of these hills, remains of glacis deposits (G1) are documented, which are correlated with the glacis level of Carabanchel Alto (677 m) on the right bank of the Manzanares (Goy et al., 1989).

The sediments documented during the excavation of the Cantera Vieja site are found at the headwaters on the left bank of the ancient Murmullo Stream, but they do not correspond to the valley bottom or alluvial plain. These deposits do not even present a direction of current concordant with that of the stream itself, since the stream would flow towards the southeast; instead, the deposit sequences of the site present an inclination towards the northwest and west, to alluvial and colluvial deposits associated with lateral contributions from the southwestern slope of the Alto del Espinillo (Fig. 5). The accumulation of alluvial and colluvial deposits seems to be associated with the formation of a small sedimentary basin towards the northwestern edge of the deposit. This small basin has been generated by the karstic collapse processes produced in the underlying gypsum of the Miocene Lower Unit. These deformations are very frequent in the Quaternary and Miocene deposits of the MJIP.

The stratigraphy of the Pleistocene units at Cantera Vieja has been divided into three general sections containing up to seven units, each defined according to their main sedimentary characteristics. With the addition of the Miocene materials of the geological substrate, the following units have been documented from bottom to top of the sequence:

MIOCENE DEPOSITS

These are composed of two levels. The first (M1) is made up of light green to grayish clays with a high percentage of trellises and carbonate cementations that give it a white tone. The level is about 70 cm thick and has a dip to the west of 20°–30°. The second level (M2) signifies the edaphic alteration of level M1, presenting a light colour due to the greater presence of carbonate cement. The two levels belong to the Miocene Intermediate Unit (Fig. 6).

QUATERNARY (PLEISTOCENE) DEPOSITS

As mentioned above, the Pleistocene materials are divided into three main sections, each with different units (Figs. 6 and 7):

Lower or basal section:

This section is composed of two units corresponding to distinct sequences.

● Unit I

This sequence is composed of three subunits that present a high percentage of carbonate cementation of edaphic origin, which gives it a generally whitish tone and may suggest that it represents another altered section of the Miocene substrate. However, a thin level of lag-like sands in its basal zone (unit I.1) and the presence of prismatic columnar structures in the upper section (unit I.3) identify this unit as part of the Quaternary sequence. The unit is an alluvial deposit at the foot of the hill, specifically a glacis type formed courtesy of a gentle slope that has undergone strong edaphic formation and subsequent weathering. It appears to be an ancient Pleistocene deposit that may correspond to the southwestern foothills of the glacis (G1) of Alto del Espinillo.

● Unit II

These are medium to coarse sands with a silty matrix and some scattered flint pebbles (unit II.1). Its thickness varies between 10 and 25 cm due to the erosive action of the upper unit. It should be considered as a glacis-type alluvial sequence distinct from that of unit I. Luminescence dating sample CVJ18-2 was taken from this unit along the southern profile. At this same level, and at the top of unit I.3, micromorphology sample CV1 was also taken, which returned a certain selection of grain sizes and some roundness in the majority-feldspar grains, thus confirming unit II represents sheet flood facies in an alluvial environment. At the western end of the excavated site, whitish silts with small, soft pebbles (0.5 cm) of brown clays have been documented, possibly from the partial dismantling of the Btk horizon of unit I.3, thus assigning this basal level of whitish silts with soft edges to unit II.0 (Fig. 6).

Intermediate section:

In the intermediate section, the levels are more similar to alluvial-colluvial sequences formed by processes involving greater or lesser drainage energy. This section is characterized by the presence of lags or alignments of pebbles that have different sizes and different lateral continuity, as well as being arranged with an increasingly pronounced inclination to the west. Most of the layers of small flint pebbles found at the base of these small sequences have numerous remains of Palaeolithic industry. Three main sequences, or units, have been differentiated.

● Unit III

This unit is made up of three sections of a predominantly colluvial sequence, or one belonging to the most apical area of an alluvial fan, where debris flow processes are also prevalent.

- Unit III.1. This level is documented along the original southern profile of Ramal 4. It is composed of sands with a high percentage of silt, and a somewhat clayey matrix. It has a dark brown hue is relatively thin (3–5 cm) and exhibits lateral continuity.

- Unit III.2. This deposit begins with a level of flint boulders of different sizes that become more voluminous (in excess of 0.5 m) towards the west. These blocks are located in the eastern zone of the excavation (grid squares F, G, H and I). No archaeological remains have been found in this level. The increase in size of the flint boulders occurs abruptly after the slope of the underlying level equally increases (perhaps caused by a pulse associated with a small karst collapse). Depending on the size of the blocks, this level has a thickness that varies between 20 cm and 60 cm towards the west-central zone.

- Unit III.3. A layer of dark brown fine sediment overlies the boulder level, and is associated with a loss of energy in the formation of the deposit. These are medium sands with an abundant silty matrix that becomes slightly more clayey towards the top. The deposits also contain some isolated flint pebbles. The full depth of this level can be observed when the accumulation of flint blocks abruptly disappears towards the west, reaching up to 60 cm in the western zone.

● Unit IV

This sedimentation model can be assigned to the sequences of archaeological levels 2A and 2B, which unit IV incorporates. The small sequences that make up this unit appear to have more of an alluvial character, with a predominance of sheet floods. In general, these sequences are formed sporadically due to strong rainfall episodes that generate surface runoff, occasioning only a minimal ‘dragging’ of the coarser, pre-existing elements from the top of the previous sequence or episode, which would preserve the archaeological levels.

This sedimentation model can be assigned to the sequences of archaeological levels 2A and 2B. It is necessary to take into account the lateral and prograding accretion of the small alluvial bodies that cover the previously accumulated deposits, and there may be several such sequences within Unit IV. For simplicity, two general sequences composed of two units each have been described.

- Unit IV.1. This is a level of aligned and at times somewhat scattered pebbles, composed of flint pebbles, flint lithic industry and debris, corresponding to archaeological level 2B. It has a slightly clayey/silty matrix and represents the basal portion of Unit IV. The layer dips towards the west in an undulating pattern.

- Unit IV.2. This is the upper section of the sequence that makes up units IV.1 and IV.2. It is composed of slightly clayey silts with a sandy fraction and can have a whitish colour due to the presence of carbonate cementation. The thickness of this unit is variable and generally thin, not exceeding 5 cm, except at the western end, where its thickness can reach 10 cm.

- Unit IV.3. This level contains an accumulation of aligned pebbles, composed of flint pebbles, flint lithic industry and debris, which is assigned to archaeological level 2A. The level has a brown silt-clay matrix and is thin (3–5 cm) but maintains a certain lateral continuity and is arranged in an undulating geometry dipping slightly towards the west. It has also been found in the initial southern profile of the excavation area and throughout the excavation area. In some areas of the excavation, unit IV.3 (level 2A) is placed above and in contact with unit IV.1 (level 2B).

- Unit IV.4. This is the upper portion of the sequence made up of units IV.3 and IV.4. It is composed of some clayey silts with a sandy fraction and has a dark brown colour (Munsel HUE-10YR). The thickness of this unit is variable, not exceeding 10 cm, while also maintaining a certain lateral continuity. These somewhat clayey brown silts are characteristic of Unit IV, where the archaeological levels are located. In this unit, luminescence dating sample CVJ18-1 was taken.

● Unit V

This unit is composed of two sequences displaying a more pronounced alluvial origin, in which the fine detritic material becomes sandier in comparison to the underlying sequences of unit IV. It contains two of the Cantera Vieja archaeological levels (level R and level 1) and coincides with a higher percentage of eroded lithic industry.

- Unit V.1. This level is an accumulation of pebbles and poorly graded sands (debris flow) with an undulating geometric arrangement. It is composed of small- to intermediate-sized flint pebbles, and lithic industry transported and eroded by environmental processes corresponding to archaeological level R. The level features a silty matrix that is lighter-coloured than the underlying levels. Its thickness is limited (2–4 cm), and it is usually found overlying and in contact with unit IV.3 (archaeological level 2A), thus maintaining a certain lateral continuity and also appearing to dip slightly towards the west. It has also been observed in the initial southern profile of the excavation in a clearly linear distribution across the excavated area. This level was chosen to characterize the CV2 micromorphology sample, which reaffirmed the poorly sorted and rolled character of the grains from this level, being interpreted as a debris flow process within an alluvial environment.

- Unit V.2. This is the upper part of the sequence composed of units V.1 and V.2. It is made up of silts and has a beige colour. The thickness of this unit is limited, not exceeding 5 cm, and it also maintains a certain lateral continuity.

- Unit V.3. This level is the uppermost accumulation of aligned pebbles, composed of small- to medium-sized flint pebbles and rolled and fresh lithic industry, which is assigned to archaeological level 1. It has a beige silty matrix, a thickness of 4–6 cm and occasionally appears above and in contact with unit V.1 (level R). It has a certain lateral continuity and also appears to dip slightly to the west. This unit was found in practically the entire excavation area.

- Unit V.4. This represents the upper section of the sequence made up of units V.3 and V.4. It is composed of silts with a certain percentage of medium and coarse sand and some clay. It has a beige colour, and its thickness varies between 50 cm in the west and 30 cm in the east of the excavation. Towards the base, occasional stones or pieces of lithic industry appear, constituting the upper part of archaeological level 1, and it also dips slightly towards the west. Luminescence dating sample CVJ18-3 was taken from the southern profile of this unit.

Upper section:

The units of this section have characteristics associated with gently sloping alluvial deposits of piedmont glacis and are affected by subsequent and strong edaphic processes.

● Unit VI

This unit is composed of a predominantly sandy glacis-like sequence with little evidence of heavily rolled pieces. It consists of a section of very coarse sand (VI.1) with millimetric-thick silt intercalations. The top of the unit has a more clayey section with the presence of carbonate capping, forming a Btk edaphic horizon (VI.2).

● Unit VII

This glacis-type alluvial sequence caps the geological deposits around the site. It consists of a 35 cm thick basal section of somewhat silty sands (VII.1) that transition to a brown silty clay with progressively less sandy fractions, constituting a strong Bt edaphic horizon with prismatic columnar structures (VII.2).

7.1 Sedimentary model and micromorphological interpretation

Study of the CV1 sample section suggests that the area was flooded with low-energy waters, although in some cases the energy would have been stronger, leaving sandier facies (see supplementary information figures. S6E and F). The low roundness of the grains and the high amount of feldspars, such as granite lithoclasts, suggest relatively short transport distances, and indicate that the deposits may be derived from other reworked, upstream, older Quaternary sediments (glacis). They have been interpreted as alluvial flood deposits. Results of CV2 thin sections (see Supplementary Information figure S7) show a poorly sorted, heterogeneous nature deposit, therefore it is interpreted that the levels were formed from mass movements (e.g. debris flow), which would have resedimented older deposits located in the vicinity. The lower plagioclase content compared to potassium feldspar also suggests resedimentation of surrounding deposits, where weathering has had a more significant impact on the plagioclase fraction (Folk, 1974; Nesbitt et al., 1997). They have been interpreted as colluvial deposits with mass movement.

With all the geological and stratigraphic data collected from field and desk work, an interpretation of these data has been made by elaborating a sedimentary model, which is synthesized with a geological section or profile. According to this model, the Miocene and Quaternary layers documented at the Cantera Vieja site present a general inclination towards the west that may be related to a progressive unconformity that affects the two layers of materials. Since both present this progressive dip, they are possibly conditioned by some karst collapse produced in the western zone, adjacent to or underlying the excavated area (Fig. 8).

It should be supposed that the geological origin of these deposits is associated with sedimentation of slope deposits towards the foot of a slope or lower down, where there is an abundance of small alluvial bodies (40–100 cm wide) and between which they also overlap laterally. Accordingly, in the initial southern profile and in the other profiles of the site, it has been possible to document numerous alluvial sequences (of small dimensions) composed of an alignment of small basal pebbles and a silty fill towards the top. The thickness of these alluvial sequences can vary between 5 and 20 cm; their longitudinal continuity extending from the slope can also vary between 20 and 100 cm, overlapping laterally (frontal or prograding accretion) to the west. However, the sandier or more clayey composition of the silts that constitute the fine fraction of the sequences has enabled the differentiation of these alluvial bodies into two stratigraphic units: unit IV (where archaeological level 2 is located) has a somewhat more clayey composition than unit V (where level R and archaeological level 1 are located). At the same time, the degree of conservation (degree of transportation/freshness) of the lithic industry remains also serves to differentiate the archaeological levels, especially when they are joined laterally or joined together in the absence of the silty section at the roof top of the mini-sequences. The study of the micromorphological sections confirms most of the geological interpretations outlined in the previous section.

7.2 Dating results

The results of the luminescence dating dose rate evaluations are summarized in Table S1 and Table S2 (see supplementary material). According to the results obtained from the HRGS measurements, the ²³⁸U and ²³²Th chains in these samples exhibit present-day secular equilibrium (i.e. the daughter–parent isotopic ratios for ²³⁸U, ²²⁶Ra, ²¹⁰Pb, ²²⁸Ra and ²²⁸Th are consistent with unity at either 1σ or 2σ; Table S2).

The SAR (single-aliquot regeneration dose) protocol shown in Table S3 (see supplementary material) was used to measure 900 grains of sample CVJ18-1 and 1800 grains of sample CVJ18-3 for TT-OSL D_e determination. The proportion of grains that passed the SAR quality assurance criteria are summarized in Table S6. Both samples exhibit relatively low proportions of TT-OSL-producing grains, with > 80% of measured grains rejected because of non-detectable or low signal intensities. The proportion of grains that passed the SAR rejection criteria and were considered for calculating the final D_e estimation was 8.4% (n = 76) for sample CVJ18-1 and 4.6% (n = 83) for sample CVJ18-3.

In general, aliquots/grains that passed the D_e measurement quality assurance criteria exhibit suitable luminescence characteristics for dating purposes, with dose–response curves capable of determining finite D_e values over the natural burial dose ranges of all samples. The pIR-IR₂₂₅ decay curves (Figure S1A) typically decrease by ~ 90% within the first 30 s of stimulation, and all the D_e values were obtained from the region of the pIR-IR₂₂₅ dose–response that was not in saturation (characteristic saturation dose or D₀ values are typically ~ 500 Gy). The TT-OSL signals (Figure S1B) typically show fast decay rates (decaying down to background levels within 0.5 s of stimulation) consistent with charge transfer into the most readily bleached (so-called ‘fast’) OSL dating trap, and the TT-OSL dose–response curves exhibit high saturation dose limits (characteristic saturation dose or D₀ values are typically > 500 Gy).

Laboratory assessments of pIR-IR₂₂₅ athermal signal stability were performed on all three samples according to the procedures detailed in the Supplementary Information. The weighted average anomalous fading rate (g-value; normalized to 2 days) for these samples is 1.86 ± 0.19%/decade (Table S9), though the values were systematically higher for sample CVJ18-2 when compared to samples CVJ18-1 and CVJ18-3 (~ 2.3%/decade vs ~ 1.5%/decade). This average empirical fading rate is consistent with published g-values for higher-temperature pIR-IR signals (e.g. pIR-IR₂₉₀; see summary in Arnold et al., 2015), as well as for athermally stable quartz OSL signals (Buylaert et al., 2012), and does not provide strong support for pIR-IR₂₂₅ age correction (see discussions in the Supplementary Information). We therefore favour the uncorrected pIR-IR₂₂₅ ages for final chronological evaluations.

A summary of the pIR-IR₂₂₅ and single-grain TT-OSL D_e values, overdispersion and final ages for the Cantera Vieja sediment samples is shown in Table 1, with the more detailed luminescence dating results shown in the Supplementary Information (Tables S7–S8). The D_e distribution of each sample is depicted graphically as radial plots in Figures S2 and S3. The pIR-IR₂₂₅ D_e datasets of all three samples exhibit low to moderate overdispersion values, which range between 4% and 19%, and are generally indicative of single dose populations. Additionally, the weighted skewness test results for these D_e distributions do not indicate the presence of statistically significant tails of high D_e values that might otherwise be characteristic of partial bleaching (Bailey and Arnold, 2006; Arnold et al., 2007). Given these characteristics, we have derived representative burial doses for the pIR-IR₂₂₅ D_e datasets using the central age model (CAM) of Galbraith et al. (1999) (Table 2; Table S7; Figure S2).

The single-grain TT-OSL D_e distribution of sample CVJ18-1 also shows low overdispersion (23 ± 4%), consistent with (i.e. within 2σ or lower than) values typically reported for well-bleached and unmixed single-grain TT-OSL D_e datasets (e.g. the average overdispersion values of 21 ± 2% presented by Arnold et al., 2019) (Table S8; Figure S3C). The final burial dose estimate for this D_e distribution was obtained using the CAM, which was favoured over the 3- and 4-parameter minimum age models (MAM-3 and MAM-4; Galbraith et al., 1999) according to both the weighted skewness score and the maximum log likelihood scores (L_max) (Arnold and Roberts, 2009; Arnold et al., 2009) (see details in Table S8). In contrast, the single-grain TT-OSL D_e dataset of CVJ18-3 is more scattered, has a higher overdispersion value (41 ± 5%; Table 1), and the D_e distribution is significantly negatively skewed at 2σ according to the weighted skewness test (see details in Table S8). The application of the finite mixture model (FMM; Galbraith and Green, 1990) reveals the presence of two discrete dose populations in this dataset (Figure S3B). The dominant FMM component (i.e. that containing the highest proportion of individual D_e values – 91% of grains; Table S8) yields an age of 197 ± 14 ka, which is in close agreement with the comparative pIR-IR₂₂₅ age for this sample (cf. 212 ± 12 ka; Table 1). In contrast the minor dose FMM component (containing 9% of grains) yields a significantly lower D_e value of ~ 260 Gy and an age of 63 ± 11 ka, which significantly underestimates the corresponding pIR-IR₂₂₅ age by a factor of 3–4 (Table S8).

The minor low-dose component for this sample likely originates from grains that have thermally unstable TT-OSL signals (e.g. Arnold and Demuro, 2015; Duval et al., 2022) or unreliable quartz types that do not fulfil basic SAR suitability requirements (grains displaying experimentally sensitized signals or slowly bleaching signals that do not originate from genuine thermal transfer of charge into the fast OSL trap) (e.g. Bartz et al., 2019; Herries et al., 2022). We can reasonably discount post-depositional mixing as an explanation for the discrete low-dose components of this sample, given the preservation of clear sedimentary structures and boundaries between the host unit V.4 and overlying unit VI.1. Similarly, beta dose heterogeneity is unlikely to give rise to such extreme and discrete dose components in most typical sedimentary contexts (e.g. Nathan et al., 2003). Given that intrinsic rather than extrinsic sources of D_e scatter likely dominate the single-grain TT-OSL dataset of CVJ18-3, we consider the FMM suitable for final D_e determination. The high-dose FMM population (FMM2) is taken as being most representative of the true burial age for CVJ18-3 as it contains the vast majority of accepted grains (91%) and yields an age that is more consistent with the Middle Pleistocene archaeological assemblage preserved at the site.

The final ages obtained for the Cantera Vieja samples are shown in Table 1. The TT-OSL and pIR-IR₂₂₅ ages for sample CVJ18-3 (collected above archaeological level 1) are 197 ± 14 ka and 212 ± 12 ka, respectively, which are very similar to the corresponding ages of 194 ± 13 ka and 215 ± 13 ka obtained for CVJ18-1 (underlying archaeological level 1). The pIR-IR₂₂₅ age for CVJ18-2 (underlying archaeological level 2) is 215 ± 16 ka and is very similar to those obtained for level 1, possibly indicating rapid sediment deposition at the site. Though the single-grain TT-OSL ages are systematically younger than the pIR-IR₂₂₅ counterparts, the two sets of ages are within 1σ of each other. Collectively, the luminescence ages are in statistical agreement and are stratigraphically consistent and indicate that levels 1 and 2 of Cantera Vieja were deposited firmly within MIS 7.

Table 1

Final luminescence ages for the Cantera Vieja sediment samples.
Sample	Relationship with archaeological level	Signal	Mineral ^a	Total dose rate (Gy/ka) ^b	Age model ^c	Overdispersion (%)	D_e (Gy) ^b	Age (ka) ^b
CVJ18-3	Above level 1	Single-grain TT-OSL	Quartz	4.24 ± 0.22	FMM2	41 ± 5	837 ± 39	197 ± 14
CVJ18-3	Above level 1	pIR-IR₂₂₅	K-Fd	4.70 ± 0.22	CAM	4 ± 2	995 ± 21	212 ± 12
CVJ18-1	Below level 1	Single-grain TT-OSL	Quartz	4.31 ± 0.22	CAM	23 ± 4	836 ± 36	194 ± 13
CVJ18-1	Below level 1	pIR-IR₂₂₅	K-Fd	4.77 ± 0.22	CAM	7 ± 2	1028 ± 30	215 ± 13
CVJ18-2	Below level 2	pIR-IR₂₂₅	K-Fd	3.87 ± 0.18	CAM	19 ± 4	832 ± 46	215 ± 16

^a All measurements were made on the 90–125 µm grain size fraction.

^b Mean ± total uncertainty (1σ or 68% confidence interval), calculated as the quadratic sum of the random and systematic uncertainties.

^c CAM = central age model (Galbraith et al., 1999); FMM = finite mixture model (Galbraith and Green, 1990).

8.1 Spatial distribution of archaeological records

Due to the urgency of the fieldwork, we were only able to excavate a portion of the original occupation surface. However, the abundance of the archaeological record indicates a clear representative distribution of lithic categories. This random distribution confirms that the main purpose of the knapping activity was the production of bifaces, with the complementary presence of flaking processes and the possible existence of sedimentary alterations by a palaeo-stream, which generated the more organized dispersion of materials in level R compared with levels 1 and 2.

The various archaeological levels have been distinguished by the characteristics of the lithic industry present and the sedimentological features. The uppermost archaeological level, named level 1 (geological unit V.3), is identified by its silty sediment texture and highly altered lithic industry, which is attributable to post-depositional processes, that is mixed with fresh lithic industry according to a linear distribution pattern (Fig. 9).

Below this, level R (geological unit V.1) is present in a delimited area characterized by abundant eroded lithic materials, with a high frequency of pieces oriented on their vertical slopes. This significant presence of vertically oriented materials, especially towards the sides of level R, shows a twisted and aligned distribution that only extends over part of the excavation plan. It seems that this level could correspond to a palaeo-stream that reworked archaeological materials from previous levels.

Finally, levels 2A and 2B (geological units IV.3 and IV.1) present the largest quantity of lithic elements. This level contains the greatest amount of fresh lithic industry, with a residual proportion of rolled industry and a profusion of knapping residues. This industry is composed of thousands of shaping flakes and fragments mixed with natural and tested blocks of flint. The characteristics of level 2 point to the primary exploitation of some blocks that show initial exploitation marks.

The dimensional distribution of different lithic materials also confirms shared organizational trends. For example, the distribution of pieces with larger or smaller thicknesses in level 2 exhibits similar densities.

8.2 The lithic assemblage

The total fresh lithic industry analysed comprises 6352 items (see Table 2). The general categories indicate that both levels 1 and 2 correspond to similar episodes that probably formed during a discrete period of time and were eroded in a particular phase by an erosive episode that introduced some external items registered in level R. The archaeological deposit seems to be produced in the format of a palimpsest but shows similar technological objectives in both levels (1 and 2).

Table 2

Principal lithic categories at Cantera Vieja: Levels 1, 2 and R.
Category	No. pieces	%
Handaxe (stage 5)	67	1.05
Bifacial preforms (stages 2–4)	129	2.03
Handaxe distal fragments (stage 5)	5	0.08
Cleaver	1	0.02
Cortical flake (stage 1)	244	3.84
Semi-cortical flake (stage 1)	1062	16.72
Cortical/semi-cortical fragments (stage 1)	415	6.53
Flakes (stages 2–4)	2741	43.15
Flake fragments (stages 2–4)	1544	24.31
Laminar flakes (stages 2–4)	20	0.31
Cores (small debitage)	93	1.46
Hammerstones	19	0.30
Other	12	0.19
Total	6352	100%

The total of 33,015 lithic elements (6352 coordinated and 26,663 recovered in 50 × 50 cm sub-squares or sectors) represents a density of 687.8 pieces per m². The density of lithic industry varies by level and concentration. Thus, in level 1, which produced 20,692 elements (3954 coordinated and 16,738 pieces recovered in sectors), the density is 431.08 per m², decreasing in level 2 to 250.06 pieces per m² (i.e. 12,003 pieces, of which 2324 were coordinated and 9679 were recovered in sectors). Finally, the 320 elements from level R (74 coordinated and 246 pieces recovered in sectors) represent a density of only 6.66 pieces per m².

From a typological and technological perspective, the assemblage demonstrates a clear ascription to the Acheulian technocomplex. Bifaces and bifacial preforms dominate the assemblage with some minor trihedral pieces and large flakes as supports (Fig. 10). The group of flaking processes is limited due to the absence of clear discoid or Levallois concepts and the presence of cores-on-flakes.

The large number of nodules and flint flakes and fragments provide evidence for the intensity of the knapping activities that took place at the site (Figs. 11A, B). In addition, the large number of shaping flakes of different stages in the bifacial configuration suggest in situ configuration of handaxes. This intensity of knapping activities does not correspond to finished pieces but to initial and intermediate biface stages, as well as to broken finished handaxes. The presence of some recycling activities has also been documented, as is frequent for these catchment contexts (Baena et al., 2015). Finally, a minor presence of retouched pieces suggests that some consumption activities took place at the site. However, the traceological study indicates that this consumption was negligible and the global strategy occurring at the site was dominated by the exploitation and transformation of flint.

The general chaîne opératoire documented at Cantera Vieja (Fig. 12) shows a clear dominance of bifacial shaping processes and a minor component of flaking activities. Bifacial sequences present elements from different bifacial stages and a variety of morphologies initially interpreted as the result of knapping skill variability.

8.3 The debitage or flaking process

a) Initial supports for the debitage

Corresponding to the raw material acquisition phase, 187 flint blocks have been documented inside the excavated area. Most of them are nodules without intense evidence of knapping, although some pieces indicate an important testing activity. According to the spatial relation of these blocks with the worked lithic pieces and the macroscopic characterization of the flint and the cortical surfaces, it is considered that these types of supports represent the raw material exploited during the period of occupation of Cantera Vieja. The catchment activity indicates that the in-ground flint blocks were removed from the matrix in order to be tested, with a similar strategy documented at the Charco Hondo site (Bárez et al., 2016).

The large number of flint nodules exhibit heterogeneous dimensions and morphologies (Figure 11C), from small nodules of about 15–20 cm length to large blocks that reach 1 m in length. With sinuous surfaces of variable quality, this profusion of raw material makes Cantera Vieja an ideal place for the procurement of supports and their initial transformation. In the studied area, only a few examples of large flake cores have been recovered. In spite of their low abundance, their presence their presence is important to an understanding the character of the chaîne opératoire.

b) Cores and flaking systems

Of the total number of pieces documented at Cantera Vieja, flaking cores represent only 1.46% of the sample and occupy small dimensional categories (maximum length <15 cm; maximum width <8 cm), except for some giant cores.

The technological analysis shows the dominance of tested cores with between one and three negatives of detaches, and a lesser predominance of some discoid cores with unipolar and centripetal flaking modalities. At Cantera Vieja, the production of medium to small flakes seems to have been a residual activity (see figure 13), since the proportion of products related to this action is limited. However, its presence also indicates that some particular process, probably associated with local activities, such as butchering or precision tool use, took place at the site as a complement to the principal objective of the lithic knapping process.

a) Supports for the shaping process

As previously mentioned, in the Cantera Vieja assemblage the debitage, or flaking, sequences for large flake production are not as prominently represented as the later stages of the bifacial series. In some other regional Acheulian sites, as is the case at Charco Hondo 2 (Bárez et al., 2011, 2016; Baena Preysler et al., 2018), the representation of final bifaces is minimal, with a high proportion of large flakes and initial bifacial preforms. At Charco Hondo 2, this representation is also accompanied by giant and large flint cores.

At Cantera Vieja, by contrast, large flakes and large cores are limited in relation to the bifacial pieces. The existence of local flint nodule blocks and some examples of large flakes and cores i indicate that these initial procurement stages in the bifacial production took place in areas close to but not immediately at the site, with short-distance transportation of the initial flake supports (Figure 14A).

b) Large flakes for shaping

The flaking products from Cantera Vieja display a discrete presence of large flakes destined for later bifacial configuration (Table 3). In particular, the bifacial configuration is represented by a total of 4067 flaking products (omitting the fragments), of which 71.92% are simple shaping flakes, 26.61% are semi-cortical flakes, 0.88% are cortical products and 0.59% laminar flakes.

The shaping flakes correspond to all stages of the bifacial production, from the creation of edges to the final formatting. The shaping flakes of stages 2 and 3 (creating edges and initial thinning) are the most represented at Cantera Vieja, accounting for 77.68% of the products, followed by flakes of stage 3 (21.17%) and, far behind, the final formatting flakes of stage 5, which represent only 1.15% of the products. The latter possibly reflects the fact that flakes from stage 5 are the smallest products, and these were collected in the 50 × 50 cm areas with reference to each level (Table 4).

Table 3. Flaking products from Cantera Vieja.

Category	Stage
Category	2 (n=1484) (%)	3 (n=1161) (%)	4 (n=721) (%)	5 (n=39) (%)	Total (n=3405) (%)
Cortical flakes	0.76	0.12	0.00	0.00	0.88
Laminar flakes	0.15	0.29	0.15	0.00	0.59
Semi-cortical flakes	22.58	3.91	0.12	0.00	26.61
Simple flakes	20.09	29.78	20.91	1.15	71.92
Total	43.58	34.10	21.17	1.15	100.00

In the general bags of small material recovered by square and stratigraphic unit, we recovered mainly flakes, fragments and residues generated during the reshaping stages in bifacial production, with a total of 26,663 pieces from all the levels, out of which 16,738 belonged to level 1, 9679 to levels 2A and 2B, and 246 to level R.

Table 4. Small material recovered by square and stratigraphic unit at Cantera Vieja.

	Flakes and fragments <3 cm (thick)	Flakes and fragments <3 cm (thin)	Flakes and fragments <2 cm
Level 1	1685	2244	12,659
Level 2	1066	1792	6971
Level R	32	59	155
Total	2783	4095	19,785

c) Preforms and final configuration products

The Cantera Vieja assemblage is characterized by the abundance of bifacial pieces (67 in total) and preforms (129 in total), as well as fractured finished elements of this reduction sequence, such as biface tips or bases (5 pieces). The collection of bifacial elements indicates great morphological variability and includes different types, some of which are the result of recycling processes (see below). Some examples (Figures 14C, D, E, F) consist of partial bifacial and trihedral pieces, while others merit only an ambiguous typological ascription (Torres Navas, 2022) that could be the result of low-skill knapping contribution to the collective production. Some examples show clear technical and technological errors and limitations, while others correspond to high-skill knapping ability (see figures S9, S10, S11, S12 in supplementary information). One example (Figure 14B) is a bifacial preform that, from the outset, shows its infeasibility for bifacial shaping due to the incorrect selection of a cracked support.

As regards the shaping flakes and fragments, 33.33% belong to bifacial stage 5 (finished pieces). On the other hand, we document a total of 129 preforms, which equates to 64.18% of the total bifacial elements. Of these, 23 preforms correspond to stage 1 (initial phases in which the first type of extractions are tested and a mise-en-forme is carried out on the starting support). From stage 2 (the creation of edges), we draw 49 pieces, and for stages 3 and 4 (the thinning of the biface), we document 41 and 16 pieces, respectively. This stage is also characterized by the intensive use of organic/soft hammerstone percussion (Key and Dunmore, 2018; García-Medrano et al., 2023), suggested as a common technique employed in the late Acheulian (Figure 15).

As mentioned above, the final objective of activities at Cantera Vieja was the production of bifaces. The elements in the final stages of bifacial production numbered 67 almost complete bifaces and 5 biface fragments (see table 5). The differences in their morphologies have allowed us to undertake a three-dimensional morphometrical study of these bifaces to enable comparison with those from other assemblages.

Table 5. Stages in the bifacial production at Cantera Vieja.

Category	Stage
	1 (n=23) (%)	2 (n=49) (%)	3 (n=41) (%)	4 (n=16) (%)	5 (n=72) (%)	Total (n=201) (%)
Biface	0.00	0.00	0.00	0.00	33.33	33.33
Bifacial preforms	11.44	24.38	20.40	7.96	0.00	64.18
Handaxe distal fragments	0.00	0.00	0.00	0.00	2.49	2.49
Total	11.44%	24.38%	20.40%	7.96%	35.82%	100.00%

It is evident that the finished bifaces that were found fragmented were produced by high technological skill levels (Figure 16 and S10). This fact clearly suggests that the final successful pieces were taken away and placed elsewhere. This spatial segmentation of production has previously been suggested for the nearby site of Los Ahijones (Bárez et al., 2016), where the main objective of production was the selection of appropriate bifacial preforms.

8.4 Hammerstones

Nineteen hammerstones have been found in the assemblage (0.30% of the total material), mostly registered in flint and some in quartzite, all as the result of recycling items from other categories. This procurement strategy has been recorded at some other sites in the interfluvial quarries of Madrid (Baena Preysler et al., 2014; Lamas et al., 2016). Quartzite pebbles are absent in this non-fluvial sedimentary context. Their nearest occurrence is located along the Jarama River, around 5 km from the site. The local absence of appropriate supports to be used as a hammerstone could explain the existence of these targeted strategies in which pebbles were transported to the site for later use as hammerstones. The intensity of their use, attested by the fragmentation of the quartzite hammerstones and the presence of extensive activity areas across their surfaces, indicates the preferential use of quartzite as a raw material for this purpose. In other cases, exhausted flint cores and fragments were recycled for use as hammerstones. However, the preference for quartzite hammerstones is clearly attested by the lower abundance and lower intensity of use of the flint hammerstones.

8.5 Recycled lithic elements and associated categories

The technological reading of particular pieces, together with the morphological analysis of contour and shape, indicates the existence of recycling/ramification processes that also affect the ultimate count of bifacial preforms and final discarded bifaces. Apart from the hammerstones described above, two main categories have been documented: the use of bifacial shaping flakes to create retouched tools and the reshaping of major biface fragments as new bifacial tools (Figure 17).

In many cases, these biface fragments were recycled by adjusting the resulting morphology from an accidental shaping process. Many pieces (Figure S10 see Supplementary information) indicate that the process of configuration continued in the first recycled or ramified process (Figure 18), even though the initial desired morphology changed after knapping accidents, which resulted in smaller bifacial morphology.

Another recycled category consists of the creation of notches in the base of bifaces. This strategy was repeated in some pieces (Figure 19) by using the longer fragmented biface pieces in a second recycling of the ramification stage. Diacritic analysis indicates that these notches were the last step in the reuse/reshaping of the piece and could be part of a new tool production aim.

There are also examples in the assemblage of two other recycling processes: initial bifacial shaping flakes used as supports for small bifaces (Figure 20), and retouched tools.

The existence of these processes could be part of a preconceived strategy, such as the ramification proposed for particular expressions of the Mousterian (Bourguignon et al., 2004, 2006), in relation to specific functional process, or the concurrence of certain secondary sourcing and reuse carried out by non-expert knappers, which generated more random knapping behaviour.

In some cases, the general configuration process seems to have reached the finished stage, and it is possible that some bifaces were taken away from the site, in a complex, fragmented form of production (Turq et al., 2013). However, apart from this proposal, the shaping process seems to signify learning and training activities in some of the other pieces. This is confirmed by the evident existence of different skill levels in the abundant bifacial composition and the abandonment of preforms at the site with no clear traces of use. These features indicate a polyvalent logic in the use of the site.

8.6 Refits

No intensive refit study (reassembling the fragments into their former wholes) has been carried out on the assemblage since finalizing the fieldwork campaign. However, during the excavation of Cantera Vieja, some significant trace-bearing pieces were located that lead us to consider the lithic group ideal for this type of study in the future.

One of the refits produced during fieldwork is a high-skill biface and its distal tip (Figure 21). The refitted pieces were located in the same excavation square (D1) less than 1 m from each other, indicating low sedimentary mobility of the documented materials.

This close spatial connection suggests the presence of intensive shaping activities, particularly during the final stages of bifacial shaping. It is important to analyse, through traceology and experimental replication, whether these types of fractures were produced during the knapping activities, or are the result of post-depositional processes, or responses to consumption phases.

The analysis of the handaxes, preforms, and retouched pieces at Cantera Vieja points to the existence of post-sedimentary alterations to the material that could mask use, but which has nonetheless provided information on the different processes that could alter this type of material at the site. However, traces of soft material working observed in a bifacial shaping flake may indicate the processing of some types of plants (Fig. 22). The scarcity of use wear could be explained by the primary use of the site for knapping purposes but may also reflect the existence of alterations that could mask these types of traces.

Another type of wear found is knapping traces. These appear to follow patterns similar to the traces of use, and therefore the same nomenclature can be used. The presence of polishes with hard material stands out, which could be due to the use of a mineral hammer in the elaboration of tools.

Despite these very limited results, the absence of traces on the material does not necessarily mean that they were not used, but rather that the activities that took place did not leave any type of trace or that these were not preserved.

As this is an open-air deposit, post-depositional processes have affected the conservation of the material to a greater degree. In this case, the flint materials were subsequently exposed to erosive or other physical or chemical processes that altered their surfaces and structures. Some of the alterations observed are cratered topography, a consequence of the process of microtopography flattening: that is, smoothing of the raised areas of the siliceous surface. This type of alteration usually appears associated with a set of chaotic striations that are also usually accompanied by shiny, flat micropolishes that are randomly distributed over the surface due to the friction of the material with the sediment or with another lithic piece. This type of alteration has been documented in numerous pieces from Cantera Vieja (Fig. 23).

The scarcity of traces is likely to indicate that the records of traces found are not linked to clear consumption activities, suggesting that a good part of the knapping processes are related to learning and training activities in the configuration of these tools, especially in relation to differentiated skill levels (Torres, 2022). However, the presence of post-depositional processes could mask the possibility of activities leaving very shallow traces. Therefore, similar analysis of other Lower Palaeolithic assemblages from the Jarama and Manzanares interfluve would allow better contextualization of this type of site and definitive confirmation of whether this setting was a unique place for the elaboration and production of lithic tools.

On the other hand, we have detected a wide variety of bifacial lithic expressions in the assemblage. Some examples indicate different stages of bifacial production, but others show significant accumulation of technical errors and mistakes, which suggests the possible existence of asymmetries in the technological skill sets of the authors (see figures S9, S10, S11 and S12 in supplementary information), a circumstance attested in other Palaeolithic sites of this area. This clear variability in biface production at the site could be the result of two significant factors:

the existence of testing processes, which produced several mistakes and errors in different stages of biface creation; or

the concurrence of different knapping skill levels in biface production, which generated various morphologies at several stages.

We have seen in the recycled pieces that this variability also increases with the move from one stage to the next for particular bifaces. To explain the reason for this general variability, and to test whether the bifacial morphologies fit with other Acheulian regional assemblages, we combined 3D morphometric analysis with an experimental comparative collection of bifacial reductions produced by knappers of different skill levels (Fig. 24).

The 3D morphometrical comparison between Cantera Vieja (n = 49) and a limited sample of the most symmetrical and technologically organized bifaces of San Isidro (n = 3) indicates important differences between the assemblages. San Isidro is one of the most representative assemblages of the Acheulian in Madrid. This site has traditionally been assigned an age in the second part of the Middle Pleistocene, around MIS 11–13 (Santonja et al., 2001; Santonja and Pérez-González, 2010; Rubio-Jara et al., 2016) but is probably dated to a more recent period by correlation with the Jarama River sequence (Moreno et al., 2019).

We have classified the pieces of Cantera Vieja based on their technical and technological characteristics and by distinguishing between different skill levels in their configuration. By comparing the highest skill levels of Cantera Vieja with San Isidro, we have found some similarities in the pieces (see Fig. 25), which indicate the presence of high technological expertise at both sites. However, we should note that several stages of the production of bifacial products introduced some ‘morphological noise’ into the study that resulted in differences between the two sites.

The results show significant similarities between both groups. Given that San Isidro could be dated to sometime between MIS 11 and MIS 7–8 (425–240 ka) and that Cantera Vieja is firmly dated to MIS 6–7 (215–194 ka), these similarities indicate the perdurance and stasis of lithic morphologies in a regional context over several glacial–interglacial cycles (Panera et al., 2014).

It has been suggested that there were different levels of technological knapping skill present at Cantera Vieja, according to experimental comparison and technical reading (Torres Navas and Baena Preysler, 2020). The absence of complete final bifaces produced by high-level knapping skills, and the presence of complete bifaces with lower knapping abilities, suggest that the site served not only as a place to create bifaces that were transported to other functional areas but also as a training ground for novices and apprentices. The Cantera Vieja results have been analysed and compared to experimental skill data to understand this aspect further. The dendrogram of clusters at the site have been compared with experimental data and confirm that medium to low skill levels dominated the knapping activities (Fig. 26). Ovaloid morphologies were found to dominate the biface production, with a lack of elongated pieces likely due to inexperienced knapping and to recycling activities.

According to our morphometric analysis, the final biface shapes produced by high-level knapping skill at Cantera Vieja are similar to the selected sample from the San Isidro site. This raises important questions about the chronology of the Acheulian period and the persistence of technologies in this particular regional context.

The interfluvial platform of Madrid is known for its abundance of flint outcrops, which have led to the discovery of numerous archaeological sites related to flint catchment and knapping activities from different periods of prehistory. These activities were particularly intense during the Middle and Upper Pleistocene, with both Acheulian and Mousterian expressions. These periods were traditionally characterized by the presence or absence of large cutting tools like handaxes and cleavers, as well as the existence or not of Levallois flaking products. However, it is worth noting that the chronology of these sites extends until recent periods of the Upper Pleistocene.

Until now, the Cantera Vieja materials have only been interpreted as a late Acheulian assemblage on the basis of typological and technological criteria. According to its geomorphological location and these new chronology results, this site corresponds to an occupation located in a headwater area of a third-order stream, more specifically in alluvial and colluvial deposits at the foot of the slope of the northwestern end of the Alto del Espinillo. The continuous accumulation of these deposits at the foot of the slope was conditioned by the existence of a basin formed by karstic collapse of the underlying gypsiferous unit. The chronology of archaeological levels 1 and 2 covers a time interval ranging from 215 ± 16 ka for the underlying unit (II.3) of archaeological level 2 (IV.1) to 197 ± 14 ka for the overlying unit (V.4) of archaeological level 1 (V.3). That is, archaeological levels 1 and 2 of Cantera Vieja were deposited during MIS 7. According to the scant erosion suffered by the elements of the documented lithic industry, especially in level 2, it seems that the anthropic activity associated with the generation of these lithic industry remains must also be framed in these temporal intervals within MIS 7.

The Cantera Vieja lithic assemblage is characterized by an absence of standardized flaking systems and the presence of an intense bifacial shaping process organized in different technical stages, including the use of organic soft percussion. The existence of a wide array of unfinished bifacial pieces indicates the presence of complex and intense knapping activities developed by different agents. The diversity of morphologies, in many cases with mistakes and errors during the knapping sequences, together with a typological ambiguity in some of the final products, suggests the concurrence of several technical and technological skill levels in a more or less cooperative production scheme. In some examples, the shaping process continues after the appearance of raw material faults (Fig. 14B). Moreover, the absence of finished complete bifaces, in contrast with other Acheulian sites in which the accumulation of final pieces is evident (Pope et al., 2006; Walter and Trauth, 2011; Despriee et al., 2016; Méndez-Quintas et al., 2018, 2019), seems to indicate a selective transport of finished pieces. As observed, the density of bifaces, bifacial preforms, and bifacial shaping flakes indicates that the site was mainly oriented to this production. However, the residual presence of other categories, such as retouched tools on flakes, many of them as residues of the bifacial sequences, indicates that activities of daily life also took place at the site.

After examining the technical and technological skill levels used at the site via experimental comparison, we found that low-skill knapping levels tended to leave pieces in their initial stages but, in other cases, produced intensive knapping without significant change in preform morphology. At the same time, some broken symmetrical bifaces at Cantera Vieja indicate high-skill dexterity in their configuration. Equally, recycling processes introduced another source of variation and confirmed the existence of secondary sourcing of residual and discarded pieces. These circumstances suggest that knapping training and skill acquisition represent an important functionality at the site.

A comprehensive study of the materials and the chaîne opératoire also suggests the concurrence of more than one functionality at the site. The first is related to knapping learning and training activities that were produced in a context that boasted rich flint resources and provided an ideal context for training (Baena et al., 2011). This observation suggests a particular model of productive social contribution. The concurrent site use by novice knappers suggests the possible existence of particular places where experts and novices shared raw material and interacted with similar purposes in mind (training in the case of experts and learning for novices and apprentices). The existence of different degrees of working in bifaces and preforms indicates that some pieces may be attempts or simply trials. This was deduced by the recognition of different technical indicators of skill abilities in the diacritic analysis. Some particular pieces show a high degree of dexterity but with errors and breakages. This observation is consistent with the absolute absence of complete final pieces at the site and suggests that those pieces were transported out of the site.

Several noteworthy aspects characterize the assemblage at Cantera Vieja: a) the concurrence of different social skill levels in the shaping process (probably related to learning aims); b) variation in the resulting types of pieces produced, as much in their morphologies as in their styles and degree of working; and c) the existence of complex reduction sequences (ramification?), recycling, and processes that link different lithic categories and tools.

The convergence of skill levels at Cantera Vieja also demonstrates the existence of interactions between the materialization of recycling processes and, in concrete terms, the reuse of broken bifaces, as part of a complex process of knapping production and learning by re-acquisition of knapped residues. The inclusion of humans with low knapping skill levels reveals the polyfunctional significance of these catchment areas, and perhaps the contribution of learning and practising activities from children and young people.

Author Contribution

A.B. was responsible for organizing the fieldwork, coordinating the main manuscript and writing the archaeological and spatial study. C and E were in charge of writing and coordinating the micromorphological and geoarchaeological study, while D wrote and coordinated the geochronological study. F oversaw the traceological analysis. All authors collaborated on preparing the figures and reviewing the manuscript.

ACKNOWLEDGEMENTS

We thank Eduardo Mendez Quintas for his kind invitation to contribute to this issue. We also want to thank José Polo and Mayra Valenciano from ARQUEX SL for their help and generosity during the fieldwork.

This study was funded by the national project ‘On the Limits of Diversity: Neanderthal Behaviour in the Central and Southern Iberian Peninsula” 1 and 2, references PID2019-103987GB-C33 and PID2022-138590NB-C42, funded by the Agencia Estatal de Investigación (State Research Agency: AEI) and Fondo Europeo de Desarrollo Regional (FEDER). C. Torres is also a beneficiary of an FPI Grant (Formación de Personal Investigador BES-2017-079805) from the Ministry of Science and Innovation of the Government of Spain and a Juan de la Cierva postdoctoral contract (JDC2022-048779-I) with the AEI. Financial support for this research was also provided by Australian research Council (ARC) Future Fellowship Project FT200100816 and Discovery Early Career Researcher Award DE160100743 awarded to M. Demuro. The authors declare they have no financial interests.

The datasets generated by the survey research during and/or analyzed during the current study are available in the Dataverse repository https://doi.org/10.21950/O7TAFL

Alberdi, M.T., Hoyos, M., Junco, F., Lopez-Martínez, N., Morales, J., Sesé, C., SORIA, D., 1983. Bioestratigraphie et evolution sedimentaire de l’aire de Madrid. Abstract Itzterinz Coll. on Mediterranean. Neogene Continental Paleoclimatic Evolution. Montpellier 18–23.
Arnold, L.J., Demuro, M., 2015. Insights into TT-OSL signal stability from single-grain analyses of known-age deposits at Atapuerca, Spain. Quaternary Geochronology 30, 472–478.
Arnold, L.J., Roberts, R.G., 2009. Stochastic modelling of multi-grain equivalent dose (De) distributions: implications for OSL dating of sediment mixtures. Quaternary Geochronology 4, 204–230.
Arnold, L.J., Bailey, R.M., Tucker, G.E., 2007. Statistical treatment of fluvial dose distributions from southern Colorado arroyo deposits. Quaternary Geochronology 2, 162–167.
Arnold, L.J., Roberts, R.G., Galbraith, R.F., DeLong, S.B., 2009. A revised burial dose estimation procedure for optical dating of young and modern-age sediments. Quaternary Geochronology 4, 306–325.
Arnold, L.J., Duval, M., Falguères, C., Bahain, J.-J., Demuro, M., 2012. Portable gamma spectrometry with cerium-doped lanthanum bromide scintillators: suitability assessments for luminescence and electron spin dating applications. Radiation Measurements 47, 6–18.
Arnold, L.J., Demuro, M., Parés, J.M., Pérez-González, A., Arsuaga, J.L., Bermúdez de Castro, J.M., Carbonell, E., 2015. Evaluating the suitability of extended-range luminescence dating techniques over Early and Middle Pleistocene timescales: Published datasets and case studies from Atapuerca, Spain. Quaternary International 389.
Baena, J., Cuartero, F., 2006. Más allá de la tipología lítica: lectura diacrítica y experimentación como claves para la reconstrucción del proceso tecnológico. In: Maíllo, J.M., Baquedano, E. (Eds.), Miscelánea En Homenaje a Victoria Cabrera. Zona Arqueológica 7, pp. 144–161.
Baena, J., Bárez, S., Pérez-González, A., Lázaro, A., Nebot, A., Roca, M., Pérez, T., González, I., Cuartero Monteagudo, F., Rus, I., Polo, J., Márquez, R., Cabanes i Cruelles, D., Carrancho Alonso, A., 2008. El yacimiento paleolítico Cañaveral (Coslada-Madrid). La captación de recursos líticos durante el Musteriense Peninsular. Arqueoweb. Revista sobre arqueología en Internet. 9 (2), 1–32.
Baena, J., Bárez, S., Pérez-González, A., Roca, M., Lázaro, A., Márquez, R., Rus, I., Manzano, C., Cuartero, F., Ortiz, I., Rodríguez, P., Pérez, T., González, I., Polo, J., Rubio, D., Alcaraz, M., Escobar, A., 2011. Searchers and miners: first signs of Flint exploitation in Madrid’s region. In: Capote, M., Consuegra, S., Díaz-del-Río, P., Terradas, X. (Eds.), Proceedings of the 2nd International Conference of the UISPP Commission on Flint Mining in Pre- and Protohistoric Times. BAR International Series 2260, pp. 203–220.
Baena, J., Ortíz, I., Torres, C., Bárez, S., 2015. Recycling in abundance: Re-use and recycling processes in the Lower and Middle Paleolithic contexts of the central Iberian Peninsula. Quaternary International 361, 142–154. doi:10.1016/j.quaint.2014.07.007
Baena Preysler, J., Ortiz Nieto-Márquez, I., Torres Navas, C., Bárez Cueto, S., 2014. Recycling in abundance: Re-use and recycling processes in the Lower and Middle Paleolithic contexts of the central Iberian Peninsula. Quaternary International 361, 142–154. doi:10.1016/j.quaint.2014.07.007
Baena Preysler, J., Torres Navas, C., Sharon, G., 2018. Life history of a large flake biface. Quaternary Science Reviews 190, 123–136. doi:10.1016/j.quascirev.2018.04.015
Bailey, R.M., Arnold, L.J., 2006. Statistical modelling of single grain quartz De distributions and an assessment of procedures for estimating burial dose. Quaternary Science Reviews 25, 2475–2502.
Bárez, S., Pérez-González, A., 2007. Patrones de aprovechamiento minero prehistórico del Sílex de Casa Montero. In: Lario, J., Silva, P. (Eds.), XII Reunión Nacional de Cuaternario. Ávila, pp. 91–92.
Bárez, S., Rus, I., Pérez-González, A., Vega, J., 2011. Los yacimientos achelenses de “Los Ahijones”, metodología geoarqueológica y resultados preliminares de la intervención. In: Santonja, M. (Ed.), Actas de Las V Jornadas de Patrimonio Arqueológico En La Comunidad de Madrid. Los primeros pobladores: arqueología, pp. 185–200.
Bárez, S., Baena, J., Pérez-González, A., Torres, C., Rus, I., Vega, J., 2016. Acheulian flint quarries in the Madrid Tertiary basin, central Iberian Peninsula: First data obtained from geoarchaeological studies. Quaternary International 411, 329–348. doi:10.1016/j.quaint.2016.01.041
Bartz, M., Arnold, L.J., Demuro, M., Duval, M., King, G.E., Rixhon, G., Alvarez Posada, C., Parés, J.M., Brückner, H., 2019. Single-grain TT-OSL dating results confirm an Early Pleistocene age for the lower Moulouya River deposits (NE Morocco). Quaternary Geochronology 49, 138–145.
Beyene, Y., Katoh, S., Woldegabriel, G., Hart, W.K., Uto, K., Sudo, M., Kondo, M., Hyodo, M., Renne, P.R., Suwa, G., Asfaw, B., Masafumi Sudo, Megumi Kondo, Hyodo, M., Renne, P.R., Suwa, G., Asfaw, B., 2013. The characteristics and chronology of the Earliest Acheulean at Konso, Ethiopia. PNAS 110, 1584–1591. doi:10.1073/pnas.1221285110
Bourguignon, L., Faivre, J.-P., Turq, A., 2004. Ramification des chaînes opératoires: Une spécificité du moustérien ? Paléo 37–48.
Bourguignon, L., Delagnes, A., Meignen, L., 2006. Systèmes de production lithique , gestion des outillages et territoires au Paléolithique moyen : où se trouve la complexité ? In: NORMES TECHNIQUES ET PRATIQUES SOCIALES. DE LA SIMPLICITÉ DES OUTILLAGES PRÉ- ET PROTOHISTORIQUES XXVI e Rencontres Internationales d’archéologie et d’histoire d’Antibes. pp. 75–86.
Bourguignon, L., Crochet, J., Capdevila, R., Ivorra, J., Antoine, P., Agustí, J., Barsky, D., Blain, H., Boulbes, N., Bruxelles, L., Claude, J., Cochard, D., Filoux, A., Firmat, C., Lozano-Fernández, I., Magniez, P., Pelletier, M., Rios-Garaizar, J., Test, L., 2016. Bois-de-Riquet (Lézignan-la-Cèbe, Hérault): A late Early Pleistocene archeological occurrence in southern France. Quaternary International 393, 24–40.
Buylaert, J.P., Jain, M., Murray, A.S., Thomsen, K.J., Thiel, C., Sohbati, R., 2012. A robust feldspar luminescence dating method for Middle and Late Pleistocene sediments. Boreas 41, 435–451.
Calvo, J.P., Zarza, A.M.A., Cura, M.A.G. del, Ordóñez, S., Aranda, J.P.R., Montero, M.E.S., 1996. Sedimentary evolution of lake systems through the Miocene of Madrid Basin. In: Friend, P.F., Dabrio, J.C. (Eds.), Paleoclimatic and Paleohydrological Constraint. Tertiary Basins of Spain: The Stratigraphic Record of Crustal Kinematics. Cambridge University Press., Cambridge, pp. 272–277.
Carral González, P., Martín-Serrano, A., Goy, J.L., Zazo, C., 1996. Las altas suerficies del interfluvio de los ríos Manzanares-Jarama al NE de Madrid (España). Caracterización geomorfológica y edáfica. Estudios Geológicos 52.
Delagnes, A., Brenet, M., Gravina, B., Santos, F., 2023. Exploring the relative influence of raw materials, percussion techniques, and hominin skill levels on the diversity of the early Oldowan assemblages: Insights from the Shungura Formation, Lower Omo Valley, Ethiopia. PLOS ONE 18, e0283250. doi:10.1371/journal.pone.0283250
Demuro, M., Arnold, L.J., Parés, J.M., Pérez-González, A., Ortega, A.I., Arsuaga, J.L., Bermúdez de Castro, J.M., Carbonell, E., 2014. New luminescence ages for the Galería Complex archaeological site: Resolving chronological uncertainties on the Acheulean record of the Sierra de Atapuerca, northern Spain. PLoS One 9, e110169.
Demuro, M., Arnold, L.J., Parés, J.M., Sala, R., 2015. Extended-range luminescence chronologies suggest potentially complex bone accumulation histories at the Early-to-Middle Pleistocene palaeontological site of Huéscar-1 (Guadix-Baza basin, Spain). Quaternary International 389, 191–212.
Despriee, J., Courcimault, G., Moncel, M.-H., Voinchet, P., Tissoux, H., Puaud, S., Gallet, X., Bahain, J.-J., Moreno, D., Falgueres, C., 2016. The Acheulean site of la Noira (Centre region, France): Characterization of materials and alterations, choice of lacustrine millstone and evidence of anthropogenic behaviour. Quaternary International 411. doi:http://dx.doi.org/10.1016/j.quaint.2015.12.101
Díez-Martín, F., Yustos, P.S., Uribelarrea, D., Baquedano, E., Mark, D.F., Mabulla, A., Fraile, C., Duque, J., Díaz, I., Pérez-González, A., Yravedra, J., Egeland, C.P., Organista, E., Domínguez-Rodrigo, M., 2015. The Origin of The Acheulean: The 1.7 Million-Year-Old Site of FLK West, Olduvai Gorge (Tanzania). Sci. Rep. 1–9. doi:10.1038/srep17839
Duval, M., Arnold, L.J., 2013. Field gamma dose-rate assessment in natural sedimentary contexts using LaBr3(Ce) and NaI(Tl) probes: A comparison between the “threshold” and “windows” techniques. Applied Radiation and Isotopes 74, 36–45.
Duval, M., Arnold, L.J., Demuro, M., Parés, J.M., Campaña, I., Carbonell, E., Bermúdez de Castro, J.M., 2022. New chronological constraints for the lowermost stratigraphic unit of Atapuerca Gran Dolina (Burgos, N Spain). Quaternary Geochronology 71. doi:org/10.1016/j.quageo.2022.101292.
Folk, R.L., 1974. The natural history of crystalline calcium carbonate; effect of magnesium content and salinity. Journal of Sedimentary Research 44, 40–54. doi:: https://doi.org/10.1306/74D72973-2B21-11D7-8648000102C1865D Download citation file:
Galbraith, R.F., Green, P.F., 1990. Estimating the component ages in a finite mixture. Nuclear Tracks and Radiation. Nuclear Tracks and Radiation Measurements 17, 197–206.
García-Medrano, P., Moncel, M.H., Maldonado-Garrido, E., Ollé, A., Ashton, N., 2023. The Western European Acheulean: Reading variability at a regional scale. Journal of Human Evolution 179, 103357. doi:10.1016/J.JHEVOL.2023.103357
Goy, J.L., Pérez-González, A., Zazo, C., 1989. Cartografía geológica del Cuaternario, geomorfología y Memoria correspondiente de la Hoja a E. 1: 50.000 de Madrid (559).
Herisson, D., Airvaux, J., Lenoble, A., Richter, D., Claud, E., Primault, J., 2016. Between the northern and southern regions of Western Europe: The Acheulean site of La Grande Vallè (Colombiers, Vienne, France). Quaternary International 411, 108–131. doi:10.1016/j.quaint.2015.12.100
Herries, A.I.R., Arnold, L.J., Boschian, G., Blackwood, A.F., Wilson, C., Mallett, T., Armstrong, B., Demuro, M., Petchey, F., Meredith-Williams, M., Penzo-Kajewski, P., Caruana, M.V., 2022. A Marine Isotope Stage 11 coastal Acheulian workshop with associated wood at Amanzi Springs Area 1, South Africa. PLOS One 17, e0273714.
Hoyos, M., Junco, J.M., Ramirez, A., Ruíz Sanchéz, M., 1985. El Mioceno de Madrid. Geología y paleontología del terciario continental de la Provinica de Madrid 6–16.
Jiménez-Arenas, J.M., Santonja, M., Botella, M., Palmqvist, P., 2011. The oldest handaxes in Europe: fact or artefact? Journal of Archaeological Science 38, 3340–3349. doi:https://doi.org/10.1016/j.jas.2011.07.020.
Junco, F., Calvo, J.P., 1983. Cuenca de Madrid. Geologia de España Tomo 2 IGM, 534–543.
Key, A.J.M., Dunmore, C.J., 2018. Manual restrictions on Palaeolithic technological behaviours. PeerJ 2018, e5399. doi:10.7717/PEERJ.5399/SUPP-2
la Torre, I. de, 2016. The origins of the Acheulean: past and present perspectives on a major transition in human evolution. Philosophical Transactions of the Royal Society B: Biological Sciences 371. doi:10.1098/rstb.2015.0245
Lamas, V., Díaz Pérez, S., Bustos-Pérez, G., Cuartero Monteagudo, F., Torres Navas, C., Baena Preysler, J., 2016. Experimental study on flint hammerstone use in Discoid-Levallois technologies: A comparison with the workshop assemblages of the central Iberian Peninsula. Journal of Lithic Studies 3. doi:10.2218/jls.v3i2.1442
Lepre, C.J., Roche, H., Kent, D. V., Harmand, S., Quinn, R.L., Brugal, J.P., Texier, P.J., Lenoble, A., Feibel, C.S., 2011. An earlier origin for the Acheulian. Nature. doi:10.1038/nature10372
Lewis, S.G., Ashton, N., Davis, R., Hatch, M., Hoare, P.G., Voinchet, P., Bahain, J.-J., 2021. A revised terrace stratigraphy and chronology for the early Middle Pleistocene Bytham River in the Breckland of East Anglia, UK. Quaternary Science Reviews 269, 107113. doi:10.1016/j.quascirev.2021.107113
López-Recio, M., Silva, P.G.G., Tapias, F., Roquero, E., Baena, J., Carrancho, a., Arteaga, C., Morín, J., Rus, I., Villalaín, J.J.J., 2014. Geochronology and geoarchaeology of Pleistocene fluvial deposits in the Prados-Guatén Depression (Madrid Basin, Central Spain). Quaternary International 328–329, 120–135. doi:10.1016/j.quaint.2013.11.029
Manzano Espinosa, I., Tapias Gómez, F., Gorbea Pérez, M., García Martín, M.L., Agustí, E., Benito del Rey, L., Illán Illán, J.M., Forteza del Rey Oteiza, C., Millán, A., Beneitez, P., Perez, S., Fernández, M., Pérez Díaz, S., López-Sáez, J.A., 2018. El yacimiento paleolítico “Vallecas 27”: (ensanche de Vallecas, Madrid). Actas RAM 2018: Reunión de Arqueología Madrileña, 2018, ISBN 978-84-09-16074-7, págs. 37-49 37–49.
Manzano, I., Dapena, L., Expósito, A., Gómez, J., Caro, J., Álavarez, D., Roca, N., Díaz, D., Lillo, J.M., Baena, J., Debenham, N., 2011. Yacimientos paleolíticos en los Berrocales (Proyecto U.Z.P.- Desarrollo del Este de los Berrocales, Vicálvaro, Madrid). In: Santonja, M. (Ed.), Actas de Las Quintas Jornadas de Patrimonio Arqueológico En La Comunidad de Madrid. Madrid, pp. 201–204.
Méndez-Quintas, E., Santonja, M., Pérez-González, A., Duval, M., Demuro, M., Arnold, L.J., 2018. First evidence of an extensive Acheulean large cutting tool accumulation in Europe from Porto Maior (Galicia, Spain). Scientific Reports 8, 1–13. doi:10.1038/s41598-018-21320-1
Méndez-Quintas, E., Demuro, M., Arnold, L.J., Duval, M., Pérez-González, A., Santonja, M., 2019. Insights into the late stages of the Acheulean technocomplex of Western Iberia from the Arbo site (Galicia, Spain). Journal of Archaeological Science: Reports. doi:10.1016/j.jasrep.2019.101934
Méndez-Quintas, E., Santonja, M., Arnold, L.J., Cunha-Ribeiro, J.P., Silva, P.X. da, Demuro, M., Duval, M., Gomes, A., Meireles, J., Monteiro-Rodrigues, S., Pérez-González, A., 2020. The Acheulean Technocomplex of the Iberian Atlantic Margin as an Example of Technology Continuity Through the Middle Pleistocene, Journal of Paleolithic Archaeology. doi:10.1007/s41982-020-00057-2
Moncel, M.-H., García-Medrano, P., Despriée, J., Arnaud, J., Voinchet, P., Bahain, J.-J., 2021. Tracking behavioral persistence and innovations during the Middle Pleistocene in Western Europe. Shift in occupations between 700 and 450 ka at la Noira site (Centre, France). Journal of Human Evolution 156, 103009. doi:10.1016/j.jhevol.2021.103009
Moncel, M., Schreve, D., 2016. The Acheulean in Europe: Origins, evolution and dispersal. Quaternary International 411, 1–8. doi:10.1016/j.quaint.2016.08.039
Moncel, M.H., Santagata, C., Pereira, A., Nomade, S., Bahain, J.J., Voinchet, P., Piperno, M., 2019. A biface production older than 600 ka ago at Notarchirico (Southern Italy) contribution to understanding early Acheulean cognition and skills in Europe. PLoS ONE 14. doi:10.1371/journal.pone.0218591
Mosquera, M., Saladié, P., Ollé, A., Cáceres, I., Huguet, R., Villalaín, J.J., Carrancho, A., Bourlès, D., Braucher, R., Vallverdú, J., 2015. Barranc de la Boella (Catalonia, Spain): An Acheulean elephant butchering site from the European late Early Pleistocene. Journal of Quaternary Science. doi:10.1002/jqs.2800
Mosquera, M., Ollé, A., Saladié, P., Cáceres, I., Huguet, R., Rosas, A., Villalaín, J., Carrancho, A., Bourlès, D., Braucher, R., Pineda, A., Vallverdú, J., 2016. The Early Acheulean technology of Barranc de la Boella (Catalonia, Spain). Quaternary International. doi:10.1016/j.quaint.2015.05.005
Mussi, M., Skinner, M.M., Melis, R.T., Panera, J., Rubio-Jara, S., Davies, T.W., Geraads, D., Bocherens, H., Briatico, G., Cabec, A. Le, Hublin, J.-J., Gidna, A., Bonnefille, R., Bianco, L. Di, Méndez-Quintas, E., 2023. Early Homo erectus lived at high altitudes and produced both Oldowan and Acheulean tools . Science 382, 713–718. doi:10.1126/SCIENCE.ADD9115/SUPPL_FILE/SCIENCE.ADD9115_MDAR_REPRODUCIBILITY_CHECKLIST.PDF
Nesbitt, H.W., Fedo, C.M., Young, G.M., 1997. Quartz and feldspar stability, steady and non-steady-state weathering, and petrogenesis of siliciclastic sands and muds. The Journal of Geology 105, 173–192.
Ollé, A., Lombao, D., Asryan, L., García-Medrano, P., Arroyo, A., Fernández-Marchena, J.L., Yeşilova, G.C., Cáceres, I., Huguet, R., López-Polín, L., Pineda, A., García-Tabernero, A., Fidalgo, D., Rosas, A., Saladié, P., Vallverdú, J., 2023. The earliest European Acheulean: new insights into the large shaped tools from the late Early Pleistocene site of Barranc de la Boella (Tarragona, Spain). Frontiers in Earth Science 11. doi:10.3389/feart.2023.1188663
Panera, J., Rubio-Jara, S., Yravedra, J., Blain, H.A., Sesé, C., Pérez-González, A., 2014. Manzanares valley (Madrid, Spain): A good country for proboscideans and neanderthals. Quaternary International 326–327, 329–343. doi:10.1016/j.quaint.2013.09.009
Pérez-González, A., 1980. Geología y estratigrafía de los yacimientos de áridos en la llanura aluvial de Arganda (Madrid). Ocupaciones Achelenses en el Valle del Jarama 1, 15–28.
Pérez-González, A., 1994. La depresión del Tajo. Geomorfología de España 389–436.
Pope, M., Wells, C., Watson, K., 2006. Biface form and structured behaviour in the Acheulean. Lithics 27, 44–57.
Royo Gómez, J., Menéndez Puget, L., Abbad, M., 1929. Mapa y memoria explicativa de Madrid (559). Mapa Geológico de España a E. 1:50.000. IGME.
Rubio-Jara, S., Panera, J., Rodríguez-de-Tembleque, J., Santonja, M., Pérez-González, A., 2016. Large flake Acheulean in the middle of Tagus basin (Spain): Middle stretch of the river Tagus valley and lower stretches of the rivers Jarama and Manzanares valleys. Quaternary International 411, 349–366. doi:10.1016/j.quaint.2015.12.023
Santonja, M., Pérez-González, A., 2010. Mid-Pleistocene Acheulean industrial complex in the Iberian Peninsula. Quaternary International 223–224, 154–161. doi:10.1016/J.QUAINT.2010.02.010
Santonja, M., Villa, P., 2006. The Acheulian of western Europe. In: Goren-Inbar, N., Sharon, G. (Eds.), Axe Age: Acheulian Tool-Making from Quarry to Discard. Equinox, London, pp. 429–478.
Santonja, M., Pérez-González, A., Panera, J., Rubio-Jara, S., Méndez-Quintas, E., 2016. The coexistence of Acheulean and Ancient Middle Palaeolithic techno-complexes in the Middle Pleistocene of the Iberian Peninsula. Quaternary International 411, 367–377. doi:10.1016/j.quaint.2015.04.056
Scott, G.R., Gibert, L., 2009. The oldest hand-axes in Europe. Nature 461, 82–85. doi:10.1038/nature08214
Sharon, G., Sharon, G., 2016. The impact of raw material on Acheulian large flake production The impact of raw material on Acheulian large flake production. doi:10.1016/j.jas.2007.09.004
Sharon, G., Alperson-Afil, N., Goren-Inbar, N., 2011. Cultural conservatism and variability in the Acheulian sequence of Gesher Benot Ya’aqov. Journal of Human Evolution. doi:10.1016/j.jhevol.2009.11.012
Silva, P., Palomares, M., Rubio, F., Goy, J.L., Hoyos, M., Martín-Serrano, A., Zazo, C., Alberdi, M.T., 1999. Geomorfología, Estratigrafía, Paleontología y procedencia de los depósitos arcósicos cuaternarios de la depresión Prados-Guatén (Sw Madrid). Cuaternario y Geomorfología 13, 79–94.
Silva, P.G., 1988. El Cuaternario del sector centro-meridional de la Cuenca de Madrid: Aspectos geomorfológicos y neotectónicos. Universidad Complutense de Madrid.
Silva, P.G., Goy, J.., Zazo, C., 1988. Evolución geomorfológica de la confluencia de los ríos Jarama y Tajuña durante el Cuaternario (Cuenca de Madrid, España). Cuaternario y Geomorfología 2, 125–133.
Stout, D., Semaw, S., Rogers, M.J., Cauche, D., 2010. Technological variation in the earliest Oldowan from Gona, Afar, Ethiopia. Journal of Human Evolution 58, 474–491. doi:10.1016/j.jhevol.2010.02.005
Terradillos-Bernal, M., Demuro, M., Arnold, L.J., Jordá-Pardo, J.F., Clemente-Conte, I., Benito-Calvo, A., Díez Fernández-Lomana, C., 2023. San Quirce (Palencia, Spain): new chronologies for the Lower to Middle Palaeolithic transition of south‐west Europe. Journal of Quaternary Science 38, 21–37.
Tixier, J., Inizan, M., Roche, H., 1980. Préhistoire de la pierre taillée I. Terminologie et technologie, Circle de. ed. Valbonne.
Torres Navas, C., Baena Preysler, J., 2020. Experts Also Fail: a New Methodological Approach to Skills Analysis in Lithic Industries. Journal of Paleolithic Archaeology 3, 889–917. doi:10.1007/s41982-020-00063-4
Torres Navas, M. de la C., 2022. Aprendizaje durante el Achelense y el Musteriense: estudio tecnoexperimental de los contextos de captación del centro peninsular (yacimientos de Los Ahijones, Los Berrocales y El Cañaveral). Universidad Autónoma de Madrid.
Turq, A., Roebroeks, W., Bourguignon, L., Faivre, J.P., 2013. The fragmented character of Middle Palaeolithic stone tool technology. Journal of Human Evolution 65, 641–655. doi:10.1016/j.jhevol.2013.07.014
Vegas, R., Banda, E., 1982. Tectonic framework and Alpine evolution of the Iberian. Peninsula. Earth Evolution Science 4.
Vicente, G. De, Giner, J.L., Muñoz-Martín, A., GonzálezCasado, J.M., Lindo, R., 1996. Determination of present-day stress tensor and neotectonic interval in the Spanish Central System and Madrid Basin, Central Spain. Tectonophysics 266, 405–424.
Walter, M., Trauth, M., 2011. High concentrations of Acheulean handaxes from the Kariandusi, Gadeb and Olorgesailie prehistoric sites: reworked artifacts or in situ tool factories? Geophysical Research Abstracts EGU General Assembly 2011 Vol. 13, EGU2011-3521.
Wang XL, L.Y.., Wintle, A.G., 2006. Recuperated OSL dating of fine-grained quartz in Chinese loess. Quaternary Geochronology 1, 89–100.

No competing interests reported.

SupplementaryInformationfinal.docx

Download PDF

Editor assigned by journal
03 Apr, 2024
Submission checks completed at journal
03 Apr, 2024
First submitted to journal
31 Mar, 2024

You are reading this latest preprint version

The Acheulian site of Cantera Vieja (Madrid, Spain) and the Lower to Middle Palaeolithic transition in central Spain

Status:

Version 1

Abstract

Figures

1. INTRODUCTION

2. MATERIALS & METHODS

2.1 Dating methodology

2.2 Geoarchaeological and micromorphological study

2.3 Archaeological fieldwork and lithic analysis methodology

3. GEOLOGICAL CONTEXT

4. CANTERA VIEJA ON AN INTERFLUVE PLATFORM

5. GEOMORPHOLOGY OF THE SITE ENVIRONMENT

6. STRATIGRAPHY

7. STRATIGRAPHIC AND CHRONOLOGICAL RESULTS

7.1 Sedimentary model and micromorphological interpretation

7.2 Dating results

8. THE LITHIC ASSEMBLAGE RESULTS

8.1 Spatial distribution of archaeological records

8.2 The lithic assemblage

8.3 The debitage or flaking process

9. DISCUSSION

10. CONCLUSIONS

Declarations

Author Contribution

References

Additional Declarations

Supplementary Files

Status:

Version 1