我们对尼安德特文化（距今25-4万年），尤其是其早期阶段（距今25-13万年）知之甚少。除了石质工具、工具组合和极少数骨质工具之外，尼人的手工制品极少保留。此外保留下来的还有红赭石和黑锰颜料以及丧葬遗址（在欧洲、近东和中东地区共约40处），这种现代性出现的迹象极其简朴，几乎没有准确定年，导致我们对这群人类祖先的认识十分有限。

我们在此发表布鲁尼克尔遗址（法国西南）距地表深达330米处的整体或大部分破碎的石笋形成的环状构建的年代。石笋圈的规则几何形状、破碎石笋的排列和数处用火痕迹表明这是一处人类活动遗迹。我们对石笋圈中的石笋再生和烧骨进行了铀系同位素断代（U-Th）测试，辅以石笋圈中石笋顶部的测年，得出可靠的、可重复的定年：距今176500 ± 2100 年，此遗址成为人类活动形成的最古老的有准确测年数据的遗址之一。其336米的深度表明，这一时期的人类已经能控制地下环境，这是人类迈向现代性的重要一步。

布鲁尼克尔洞穴深处的石笋圈最近发表在《自然》杂志之上。科研团队由来自法国、比利时、中国和美国的专家组成。

布鲁尼克尔位于法国西南部(44°05 N, 1°66 E)亚奎丹盆地东南，在图卢兹省以北8公里，位于阿韦龙河流域，这一地区旧石器时代遗址密集，主要集中于旧石器时代晚期。

洞穴的入口处有两条狭窄的通道，进入十分困难。我们必须爬行三十余分钟，并边挖掘塌陷的入口才能抵达长500米的巷道， 幽闭恐惧症患者无法胜任此项工作。通过入口的过道之后，抵达一大片碎石废墟的顶部，其间杂有破碎的石板。这处洞穴位于地面以下336米，完全隔绝天光。

我们对入口处堆积开展了研究，重点探索其年代问题。铀系同位素断代的较早结果表明此处非常古老，至少是中更新世晚期（0,3-0,13 Ma）。S. Verheyden认为碎石堆积可能早于石笋圈，问题在于此处是否是人类进入洞穴的通道。

这片碎石堆积之上发现了大型更新世动物的遗骸，我们未经取样在原地研究的结果表明，其中包括：棕熊、狼、马鹿、驯鹿和山羊等。这些动物的年代尚属未知，但有可能来自中更新世。由于没有发掘，我们将对狼的遗骸进行测量，以确定他们的生物年代。熊也在巷道起点、碎石堆积底部留下了丰富的活动痕迹，包括冬眠洞穴、爪印和足迹等，

这两处奇特的石笋圈最初是探险者发现的。1995年法国的一个洞穴探险杂志Spelunca上发表了一张地图，是关于这处洞穴最早的研究。Rouzaud及其合作者辨认出了建筑材料（破碎或完整的石笋）、截取材料的痕迹和方解石的再生。他们同时也在石笋圈中发现了一片烧火痕迹及烧骨。Hélène Valladas通过碳十四加速质谱仪得出了至少距今4.7万年的数据，但这个数据显然受到了方法本身的限制。

首次发表之后，意料之中的是布鲁尼克尔遗址没有引起关注，尽管最初被判断为莫斯特文化，它的定年显然是很不清楚 。只有 M. Lorblanchet (1999), J. Clottes (2005, 2006) 和B. Hayden (2011, 2012) 等学者论及布鲁尼克尔遗址，并将其认定为旧石器时代中期，或者认为石笋圈是尼安德特人所为。

由于François Rouzaud的英年早逝，以及洞穴限入令的限制， 直到2014年才重新启动对该遗址的进一步研究。我们与Sophie Verheyden博士, Dominique Genty教授和Michel Soulier教授等学者一起研究石笋圈，并将研究成果发表于2016年的《自然》杂志上。

我们的研究目标主要有二，首先，利用现代科研手段开展更精确的调查，以数据库形式更准确地描述石笋圈的所有构建元素；其次，采用如下概念指导定年：石笋顶端代表的是最老年份，而底部则是石笋圈年代的下限。石笋圈的年代应该在上下限之间。

首先，石笋圈距离洞穴入口336米，建造于全洞最大的洞穴中央。

石笋圈的材料纯粹为方解石。我们已经辨认出两大类结构：两个环状圈（A和B）和4个略小的石笋堆积（C, D, E, F）。其中最大者面积为7 x 5米，小的直径约1-2米。整个结构采用了420块石笋，估计采用了重逾2吨的石笋，因此该遗迹被命名为“洞穴制品”。

进入洞穴的其它区域绝非易事。首先需要进行详实的埋藏学分析。固然部分石笋圈淹没在水中，而部分埋藏在土里。但是从埋藏学角度来看，主要问题是方解石的出现。石笋圈的大部分都被方解石覆盖，一精细层方解石、一层硬壳以及再生，均晚于石笋圈本身。通过3D建模，我们请团队中的同事尽量清除近期方解石，尤其是再生成分，以确保获得原始结构。

此外很重要的一点是证明该结构与人类活动相关。我们计算并比较了熊的洞穴与两个石笋圈的尺寸。石笋圈的主要组成部分的方向和方位角相似，但没有明显的施密德曲线偏向。这表明这种堆积和坡度模式不可能是水流、质量流或者其它重力作用自然形成的。

至于建筑材料问题，通过初步观察，我们发现“洞穴制品”基本上完全采用单体石笋而非钟乳石，偶见流石。同时我们发现，这些石笋甚少有完整的，基本都是碎块，明显是被打碎并截断所致。半数残块都是石笋的中段，其次是末梢和根部。石笋碎块有长短两类，其各自长度均有统一标准。这些测量和测试结果强烈表明这是一处人为工程。基于大型洞穴制品的存在，石笋圈的主圈可与其它建筑部分明显区分。

主圈由一到两三层，甚至四层石笋整齐垒叠而成。有趣的是，石笋层的内侧放置较短的材料，意在支撑石笋圈。其它石笋则垂直置放于主圈旁，似乎是在等待备用。

1995年发表的第一篇研究报告只确认了三处用火遗迹，而我们发现了十八处。六个石笋结构中全都有用火痕迹和包含火烧遗存的区域。只有一处用火遗迹发现在地面上，这是因为洞内的地表大多覆盖了一层方解石外壳，只有该区域的土壤暴露在外。我们已观察到烧骨、烟火痕迹、烧过的发红或发黑且有裂痕的洞穴制品，石笋圈中还发现十几枚黑色骨骼残片。

我们对这些发红或发黑的区域进行了地磁测试，确认这些遗迹都经过加热，磁性图表明方解石层下方甚至还有三处用火遗迹。Genty教授测年取样的最大烧骨片就取自方解石层之下。另一件黑色碎片密封在所谓的“洞穴产品”和方解石再生层之间。这些遗存显然是人类加热所导致的，与分子和原子光谱分析结果一致。

石笋圈的年代问题，主要通过对石笋方解石铀系同位素断代研究解答。石笋顶端是最老年龄，底部的再生代表了石笋圈的最新年龄。

我们选取了十八件样品进行电感耦合等离子体质谱仪铀系同位素测年。另有一件流石核和三件来自覆盖着烧骨的方解石硬壳样品。测年结果优先参考的是构建石笋圈的石笋，分布在距今17.7万年和40万年之间。更有趣的是覆盖在石笋圈上的方解石壳再生的年代，其两个最老的年龄刚好落在同样的时间框架内，部分与石笋圈中最年轻的石笋年龄相当。这个年代序列表明石笋顶端与方解石再生大致同时或略老于再生。

烧骨表明了人类活动的存在，其年龄早于距今18万± 2万年。石笋圈中的流石年龄接近，表示这个时期（同位素阶段6）虽然处于冰川期，但其气候却足够温暖湿润，保证了持续的方解石沉积。这个结果表示石笋圈建于距今约17.6万年。

石笋圈的建造需要选择特定的原材料（石笋），根据标准尺寸进行取材、破碎，并按计划运输及安置。所有这一切都需要一定程度的建筑技能，例如在两排构件之间插入支撑部件等。建筑过程中使用的“洞穴制品”的数量（420件）及其总重量（2.1-2.4吨）表明建筑过程简单而漫长。这项工作还需要相当的照明，考虑到进入洞室的通道较长，这必定是一项集体工程。而完成这样一个工程，需要有一个社会组织，从而进行设计、商讨和分工。

布鲁尼克尔遗址对于早期尼安德特人研究的突出重要性主要体现在以下三个方面：

其一，它揭示了包括尼安德特人和早期尼安德特人在内的前现代人种对喀斯特地下深处洞穴的利用情况。这是目前为止这个年代的此类深穴环境遗址首次得到确认。此前发现的零星几处中旧石器时代遗址要么无法得到证实，要么只包含可见天光的入口处。目前最早的证据发现在欧洲（例如约距今4.2-3.2万年的西班牙埃尔·卡斯蒂洛和法国 科维洞穴遗址）、东南亚/巽他、华莱士（距今约4-3.5万年的印度尼西亚苏拉威西Pettakere洞穴遗址）。这些遗址都不早于距今4万年，通常认为与智人相关，智人造访这些洞穴的目的在于其象征性与岩画，这实际上是另一个世界。

其二，它表现了此前从未被报导过的精密建筑活动。除法国尼斯的Lazaret洞穴遗址（距今13-16万年）之外，大多数欧洲的此类遗址都是开放场地，属于居住建筑。例如荷兰Maastricht-Belvédère，乌克兰的Molodova I，罗马尼亚的Ripiceni-Izvor和法国的La Folie等遗址，主要都集中于中旧石器时代晚期。在此我们不准备讨论布鲁尼克尔遗址的功能，因为由Brian Hayden教授主持的这方面研究才刚刚起步。

其三，发现了保存火的遗迹。中旧石器时代早期的用火遗迹十分罕见，布鲁尼克尔洞穴中应该使用过可移动的灯或者火把，但是我们也知道，考古发现最早的灯不早于旧石器时代晚期。

我们的发现表明，布鲁尼克尔遗址的尼安德特人已经具备了某些现代性要素，其出现时间远早于此前所知，早于旧石器时代晚期使用石笋也十分引人注目。远离入口的石笋圈表明尼安德特人已经掌控了这种奇特而危险的环境，我们认为这是人类历史上重要的一步。

Very little is known about Neanderthal cultures (250 000-40 000 years BP), especially early ones (250 000-130 000 BP). Other than lithic implements, lithic toolkits and exceptional bone tools, very few of their artefacts have been preserved. While those that do remain include red (ochre) and black (manganese) pigments and burial sites (~50 for Europe, Near and Middle East), these indications of modernity are extremely sparse and few have been precisely dated, thus greatly limiting our knowledge of these human ancestors.

Here we present the dating of annular constructions made of entire and mainly broken stalagmites found deep in Bruniquel Cave (SW-France) at 330 meters from the actual entrance. The regular geometry of the stalagmite circles, the arrangement of broken stalagmites and several traces of fire demonstrate the anthropogenic origin of these constructions. Uranium-series dating (U-Th) of stalagmite regrowths on the structures and on burnt bone, combined with the dating of stalagmite tips in the structures, give a solid and replicated age of 176 500 ± 2 100 years before present (BP), making these edifices among the oldest known well-dated constructions made by humans. Their presence at 336 meters from the actual entrance of the cave indicates that humans from this period already mastered the underground environment, which can be considered as a major step in human modernity.

These constructions found deep in Bruniquel Cave have been recently published in Nature (Jaubert et al., 2016) with Dr. Sophie Verheyden (Royal Belgian Institute for Natural Sciences, Brussels, Belgium), Pr. Dominique Genty (LSCE, UMR 8212 CNRS-CEA-UVSQ, Gif-sur-Yvette, France) and Michel Soulier (Speleological & Archaeological Society of Caussade, France).

The scientific team includes different specialties with the French, Belgian, Chinese and American specialists of:

–       U-Th dating: Dr. Hai Cheng (Institute of Global Environmental Change, Xi’an Jiaotong University, Xi’an, China) and Pr. R. Lawrence Edwards (Department of Earth Sciences, U. Minnesota, Minneapolis, USA),

–       Palaeoclimatology: Pr. Dominique Blamart, Édouard Régnier (LSCE, UMR 8212 CNRS-CEA-UVSQ, Gif-sur-Yvette, France),

–       Karstology: Dr. Serge Delaby (University of Mons, Belgium),

–       Topography: Dr. Hubert Camus (Protée Expert, Sommières, France), Xavier Muth (Get in Situ, Switzerland),

–       Magnetism: Dr. François Lévêque (University of La Rochelle, LIENSs, UMR 7266 CNRS, France),

–       Geoarchaeology: Dr. Catherine Ferrier (University of Bordeaux, PACEA, UMR 5199 CNRS-UB-MC, France),

–       Geophysics: Dr. Christian Burlet (Royal Belgian Institute for Natural Sciences, Brussels, Belgium),

–       Burnt remains: Damien Deldicque, Pr. Jean-Noël Rouzaud (Laboratory of Geology École Normale Supérieure, UMR CNRS 8538, Paris),

–       Ichnology: Frédéric Maksud (Ministry of Culture, Regional Archaeological  Service of Occitanie, Toulouse, France),

–       Photogrammetry and 3D modelling: Pascal Mora (Archéostransfert, Archéovision, UMS 3657 SHS-3D, Pessac, France), Xavier Muth

–       GIS: François Lacrampe-Cuyaubère (Archéosphère, Quirbajou, France)

–       Statistics: Frédéric Santos (PACEA, UMR 5199 CNRS-UB-MC, University of Bordeaux, France).

Bruniquel is located in Southwestern France (44°05 N, 1°66 E), southeast of Aquitaine Basin (Quercy), and 80 km north of Toulouse in the Aveyron valley, an area already rich in Palaeolithic sites mainly attributed to the Upper Palaeolithic.

The actual entrance is narrow and quite difficult, with two very narrow passages. We must crawl for over 30 meters. Spelunkers dug through the collapsed porch, a path through which non-claustrophobic persons can reach the gallery. The cave consists of 500 meter long gallery. After the entrance corridor, we arrive at the top of a scree (a slope with loose stones) with a large fan of debris and collapsed stone slabs. The structures are located at 336 meters from the actual entrance and daylight.

We are studying the question of the filling of this entrance, and especially its dating. Early results from speleothems (U-Th) indicate a rather old age, at least at the end of the Middle Pleistocene (0,3-0,13 Ma). According to S. Verheyden, it is likely that this scree (eboulis) may be older than the structures (cf. infra). But the question is whether humans entered into the cave through this porch.

On this scree the remains of large Pleistocene fauna (brown bear, wolf, red deer, reindeer, ibex and others) were found and studied in situ without taking samples. This fauna is not yet dated but is probably from the Middle Pleistocene. Without any excavation, we will measure the remains of the wolves in order to specify their biochronology (Dr M. Boudadi-Maligne, PACEA, UMR 5199 CNRS-UB-MC, University of Bordeaux, France). Bears also left numerous traces of their presence: hibernation hollows, claw marks and footprints are found at the beginning of the gallery, at the foot of the scree.

The first speleologists advanced only on a materialized pathway to avoid damaging the floors with modern round-trips. The concretions are splendid, with stalagmites, stalactites, draperies, lakes with floating calcite, etc. It is important to note this because Neanderthals had essentially seen the same environment.

The strange arrangement of two annular structures made of stalagmites was at first identified by the discoverers. A first study provided a map published in 1995 in Spelunca, a French journal of speleology. Rouzaud and collaborators distinguished the construction elements (broken stalagmites and whole ones), traces of extraction and the calcite regrowths. They also discovered in the structure a burnt bone in a fireplace. A single C14 AMS dating by Dr. Hélène Valladas yielded a first age of more than 47 000 BP, the limit of the method (Rouzaud et al., 1995).

After the first publication, with rare exceptions (Balter 1996, Rouzaud 1996, 1997), there was no mention of Bruniquel. Only M. Lorblanchet (1999), then J. Clottes (2005, 2006) and finally B. Hayden (2011, 2012) mentioned Bruniquel and evoked the possibility of attribution to Middle Palaeolithic or Neanderthal for the structures.

The premature death of François Rouzaud, along with restricted access to the cave, prevented any further research until 2014. With Dr. Sophie Verheyden, Pr. Dominique Genty and Michel Soulier, we decided to study these enigmatic constructions and we got the amazing results published in Nature (Jaubert et al., 2016).

We had two main goals. First, to conduct a more accurate survey with modern methods and to more accurately describe structures by identifying each element in a database form. Second, to date these structures according to a simple concept: the top of stalagmites that are part of the structure give maximum ages while the bases of the stalagmite sealing the structures (regrowths) give minimum ages. The age of the structures must be interposed between the two.

At 336 meters from the entrance, the structures were built in one of the largest chambers of the cave near its center. The structures are only composed of calcite. We have identified two categories of structures: two annular ones (A, B), and four smaller accumulations consisting of stacks of stalagmites (C, D, E, F). The largest one measures 7 x 5 meters, and the smaller 1-2 meters in diameter. A total of 420 pieces were used, here designated as ‘speleofacts’, comprising a weight estimated of more than 2 tons of stalagmites.

Access to the remains is not easy and requires significant taphonomic analysis. Indeed, parts of the constructions are partly under water, while others are covered with clay. But from a taphonomic point of view, the main problem is the presence of calcite. Most of the structures are covered by calcite – a fine layer, a thicker crust and regrowths posterior to the construction. As such, from the 3D model (photogrammetry P. Mora), we have asked our colleagues (F. Lacrampe-Cuyaubère, X. Muth) to remove as much recent calcite as possible, especially regrowths, to obtain a new reconstitution without the recent calcite.

It was important to prove the anthropic character of the structures. We calculated (with F. Santos) the comparisons of the dimensions of bear nests (hollows) and the two annular structures of Bruniquel. The orientation and azimuth data in the main structures are similar and do not show any preferential direction according to the Schmidt diagrams. This confirms that such orientation and slope patterns cannot be due to natural processes related to water flow, mass flows or other gravitational processes.

Concerning the building material used, our first observation was that the ‘speleofacts’ almost exclusively selected had been stalagmites but no stalactites. There are also some elements of flowstones. But the nature of the preferred elements is clear: they are almost exclusively stalagmites, and mainly single. Our study also shows that these stalagmites are rarely whole; they are fragments and were clearly broken and sectioned. Half of the elements consist of the middle part of stalagmites (mesial), then distal and proximal. The stalagmites are well-calibrated with a mean length for the large ones and another for the smaller ones. All statistical measurements and tests (F. Santos) strongly confirm an intentional construction. The main structure can be distinguished from the others by very large speleofacts.

The main annular structure is composed of one to two, three or even four superimposed layers of aligned stalagmites. Interestingly, some short elements were placed inside the layers of stalagmites to support them as wedging elements. Other stalagmites were placed vertically against the main structure in the manner of stays.

The first study published (Rouzaud et al., 1995) identified only three fireplaces. We discovered 18 of them. Traces of fire and areas with burnt remains are present in, and on, all six structures. Only one is on the ground. But this is normal: a thick calcite crust prevents access to the original soil, except in this zone without calcite. We have observed burned bones, soot and burned speleofacts, reddened and blackened and more or less fissured. A dozen black bone fragments are observed in the structures.

Most of these fireplaces were clearly heated, as confirmed by magnetic measurements for the most striking and evident reddened and blackened zones. The map of magnetism (F. Lévêque) reveals even three other zones of fire under calcite. The largest burnt bone sampled by Pr. D. Genty for dating was covered by a thick calcite. Another black fragment was trapped in a core between a speleofact and the calcite regrowth. These remains were clearly heated by humans, as indicated by molecular and atomic spectrometry (infrared, Raman:  Pr J.-N. Rouaud, D. Deldicque).

The age of the constructions has been determined by uranium-series dating (U-Th) of the stalagmite calcite (Pr Hai Cheng and Pr R. L. Edwards). The top of stalagmites that are part of the structure give maximum ages, while the bases of the stalagmite regrowths sealing the structures give minimum ages.

Eighteen samples were taken for dating by ICiPi-MS uranium-series (S. Verheyden, É. Régnier, D. Genty, S. Delaby). Four additional samples were also dated: one from a core taken in the flowstone and three from the calcite-crust covering a burnt bone. Ages obtained for stalagmites used in the construction provide a first indication. They are distributed between 177 000 years BP and more than 400 000 years BP. More interesting is of course the age of the calcite regrowths covering the structure. The two oldest ages are situated in the same timeframe. They partially cover the age of the youngest dated-stalagmite in the structure. All other ages according to this chronology show that the stalagmite-tips are contemporary to or older than the calcite regrowths.

Moreover, additional evidence for human presence is provided by the burnt bone, which is older than 180 000 ± 20 000 years BP. The age of the flowstone situated inside the structure is similar, suggesting that the climate during this period, covering isotope stage 6 (MIS 6), was sufficiently humid and warm enough to allow continuous calcite deposition despite generally glacial conditions. These results indicate that the structure was built around 176 000 years BP.

The construction of such structures involves the choice of the raw material (stalagmite), its extraction, fragmentation according to standard dimensions, followed by its transport and placement according to a plan. It supposes a minimum degree of skill with architectural techniques such as inserting wedging elements between two rows of construction. The number of ‘speleofacts’ used (420), as well as their weight (2,1-2,4 tons) requires simple but long procedures. This work would also have required adequate lighting. Combined with the long entrance to the chamber, this information indicates that it was a collective project and thus a form of social organization requiring design and discussion by several individuals, as well as a distribution of tasks.

The attribution of the Bruniquel constructions to Early Neanderthals is unprecedented in three ways.

First, it reveals the appropriation of a deep karst space by a pre-modern human species: Neanderthals and even Early Neanderthals. Until now no site of this age in a deep cave environment, a difficult environment, has been recognized. The few Middle Palaeolithic examples could not be verified or concern only entrance zones lit by daylight. The oldest evidence is known in Europe (e.g. El Castillo in Spain, Chauvet Cave in France: 42-32 000 cal. BP) and South-East Asia / Sunda, Wallacea (Lean Leang, Pettakere Cave, Sulawesi, Indonesia: 40-35 000 cal. BP). These sites are thus younger than 40 000 cal. BP and are always associated with Homo sapiens. Symbolic rock art was the main reason for these cave visits. It is another world.

Second, it concerns elaborate constructions that have rarely been reported before. Except for the discussed case of the Lazaret cave (Nice, France, 130-160 000 years BP), most of the examples come from European open-air sites which are habitat structures (e.g. Maastricht-Belvédère (Netherlands), Molodova I (Ukraine), Ripiceni-Izvor (Romania), La Folie (Poitiers, France)) mainly dated to the Late Middle Palaeolithic. Here we do not discuss the function of the structures of Bruniquel, a topic on which we are just beginning to work, notably with Pr. Brian Hayden.

Third, we have evidence of fire maintenance. Traces of fire are rare for the early Middle Palaeolithic. So, in Bruniquel, mobile lamps or torchlights were also forcibly used. And it is doubly interesting because, here too, the oldest archaeological lamps known are found in the Upper Palaeolithic.

Our findings suggest that the Neanderthal society of Bruniquel included elements of modernity which can now be proven to have emerged earlier than previously thought. The use of stalagmites is also unique for periods older than the Upper Palaeolithic. The presence of this kind of structure far from the entrance indicates that Neanderthals had already mastered this strange and dangerous environment, and this can be considered as an interesting step in human history.