Sonderdruck aus Archäologisches Korrespondenzblatt Jahrgang 48 · 2018 · Heft 3 Herausgegeben vom Römisch-Germanischen Zentralmuseum Mainz Leibniz-Forschungsinstitut für Archäologie REDAKTORINNEN UND REDAKTOREN Paläolithikum, Mesolithikum: Martina Barth · Harald Floss Neolithikum: Doris Mischka · Johannes Müller Bronzezeit: Christoph Huth · Stefan Wirth Hallstattzeit: Markus Egg · Dirk Krausse Latènezeit: Rupert Gebhard · Sabine Hornung · Martin Schönfelder Römische Kaiserzeit im Barbaricum: Matthias Becker · Claus von Carnap-Bornheim Provinzialrömische Archäologie: Peter Henrich · Gabriele Seitz Frühmittelalter: Brigitte Haas-Gebhard · Dieter Quast Wikingerzeit, Hochmittelalter: Hauke Jöns · Bernd Päffgen Archäologie und Naturwissenschaften: Felix Bittmann · Corina Knipper · Thomas Stöllner Die Redaktorinnen und Redaktoren begutachten als Fachredaktion die Beiträge (peer review). Das Archäologische Korrespondenzblatt wird im Arts & Humanities Citation Index® sowie im Current Contents® / Arts & Humanities von Thomson Reuters aufgeführt. Beiträge werden erbeten an die Mitglieder der Redaktion oder an das Römisch-Germanische Zentralmuseum, Ernst-Ludwig-Platz 2, 55116 Mainz, korrespondenzblatt@rgzm.de Die mit Abbildungen, einer kurzen Zusammenfassung und der Anschrift der Autorinnen und Autoren versehenen Manuskripte dürfen im Druck 20 Seiten nicht überschreiten. Die Redaktion bittet um eine allgemein verständliche Zitierweise (naturwissenschaftlich oder in Endnoten) und verweist dazu auf folgende Richtlinien: http://web.rgzm.de/publikationen/verlagsprogramm/zeitschriften/archaeologischeskorrespondenzblatt.html Die Beiträge des Archäologischen Korrespondenzblatts stehen ein Jahr nach Erscheinen der Printausgabe auch Open Access zur Verfügung und sind dank einer Kooperation mit der Universitätsbibliothek Heidelberg unter folgendem Link abrufbar: http://journals.ub.uni-heidelberg.de/index.php/ak/issue/archive ISSN 0342-734X Nachdruck, auch auszugsweise, nur mit Genehmigung des Verlages © 2018 Verlag des Römisch-Germanischen Zentralmuseums Redaktion und Satz: Manfred Albert, Marie Reiter, Martin Schönfelder Herstellung: AC medienhaus GmbH, Wiesbaden Das für diese Publikation verwendete Papier ist alterungsbeständig im Sinne der ISO 9706. PhILIPPE LEFRAnC · AnThOnY DEnAIRE a new Model foR the inteRnal oRGanization of lBk settleMents: the site of BischoffsheiM (dép. Bas-Rhin / f) and the »oRthoGonal Model« The long prevailing model for the internal organization of Linear Pottery culture (LBK) settlements theorized by the Aldenhoven plateau team – the Hof­ platz or independent homestead model – has recently been subjected to drastic criticism from a methodological point of view and a new paradigm has been proposed, claiming to replace it: the Zeilen­ siedlungsmodell or row model. We think that both these two interpretive constructions are seriously weakened by the fact that they have largely neglected important aspects of chronology, on the one hand concerning the methods of indirect dating of the houses from Langweiler (Kr. Düren / D), and on the other even completely ignoring chronological data in the case of the row model. We propose here a new model for the spatial organization of LBK villages; it is based on data obtained by the excavation of the Alsatian site of Bischoffsheim (dép. Bas-Rhin / F) and its construction is fundamentally supported by the chronological data. The Bischoffsheim LBK village unearthed in 2002 1 is located 20 km southwest of Strasbourg at the foot of the vosges mountains in the Bas-Rhin department (fig. 1). The plans of 39 houses came to light, most of them well preserved and the majority of them precisely datable thanks to the abundant material found in the construction pits of the buildings (fig. 2). fig. 1 The Bischoffsheim settlement in the upper Rhine valley (dép. Bas-Rhin / F). – (Illustration A. Denaire / Ph. Lefranc). the Method: factoRial coRRespondence analysis and aBsolute chRonoloGy The assemblages of decorated pottery found at Bischoffsheim in the lateral house pits have been integrated into a new correspondence analysis on presence-absence data of the LBK ceramic series in Lower Alsace. It Archäologisches Korrespondenzblatt 48 · 2018 307 fig. 2 Plan of the Bischoffsheim settlement (dép. Bas-Rhin / F). – (Illustration B. Bakaj / Ph. Lefranc). 308 ph. lefranc · a. denaire · A New Model for the Internal Organization of LBK Settlements fig. 3 Correspondence analysis of the LBK assemblage of Lower Alsace, with the indication of the Bischoffsheim houses (dép. BasRhin / F). – (Illustration Ph. Lefranc). relies on 140 assemblages and 139 typological criteria described by Ph. Lefranc 2. We recognize eight stylistic stages divided into four of the five big phases defined by W. meier-Arendt 3. The study of the pottery of Bischoffsheim enabled us to date precisely 31 houses out of 40 (that is to say 77 % of the corpus), ranging from stage IIB to the late stage Ivb (figs 3-4). 17 houses belong to the early LBK. Five of them contained enough material to be attributed to the IIB phase (houses 4, 7, 26, 29 and 32) and seven of them have been dated to the IIC stage (houses 1, 10-12, 27, 41 and »the village«) (figs 2-3). The five other houses can be related to either stage (houses 2, 16, 34-35 and 38). We also benefit from other dating criteria not discussed in detail here, which allow us to confirm these chronological attributions: especially the presence of the Y or pseudo Y layout of the central postholes, and also the sizes and orientations of the buildings 4. houses 3, 16, 23, 33, 36 and 40 clustered at the bottom of the right part of the ellipse in the correspondence analysis (fig. 3) are scattered in the middle of the assemblages attributed to the middle LBK. Lastly, the left line of the curve is occupied by two distinct groups: on the right, the first and more enriched one contains five late LBK houses which have provided typical materials of stage Iva1 (houses 5, 8-9, 13 and Archäologisches Korrespondenzblatt 48 · 2018 309 fig. 4 Correspondence analysis of the early and middle LBK assemblage of Lower Alsace, with the indication of the Bischoffsheim houses (dép. Bas-Rhin / F). – (Illustration Ph. Lefranc). 22); the second concerns three houses, two of which can be dated to stage Iva2 (houses 14 and 28) and another house (31) to stage Ivb. Obviously, these three assemblages are very close to each other which pleads in favour of an attribution of house 31 to the very beginning of the Ivb stage. The Bischoffsheim site was therefore probably abandoned either in the first half or perhaps at the beginning of stage Ivb. The recent Bayesian modelling of the Alsatian Danubian neolithic has enabled us to formally estimate the duration of each stylistic stage 5, expressed here in terms of generations (fig. 5). The resulting overall model suggests a colonisation of Lower Alsace taking place in the first half of the 53rd century BC and a IIB stage expanding over two or three generations. The following IIC stage covers the generation expanding in the third quarter of the 53rd century BC. The middle LBK spreads over one or two generations around the turning point between the 6th and the 5th millennia and is followed by the Iva1 style lasting for only one generation around 5160 cal BC. The Iva2 style, a rarely observed stage in the ceramic assemblages, only lasts for one generation, maybe just about ten years; it quickly leads to the Ivb stage which expands during two or three generations until 5100-5040 cal BC (at 68 % probability). Stage v (final LBK), absent in Bischoffsheim, could not be dated. 310 ph. lefranc · a. denaire · A New Model for the Internal Organization of LBK Settlements fig. 5 Stylistic stages of the Lower Alsace LBK with the indication of the corresponding houses of Bischoffsheim (dép. Bas-Rhin / F). – (Illustration Ph. Lefranc). cRitiques of the existinG Models: the Hofplatz and the zeilensiedlung Models The site of Bischoffsheim was occupied for a long time, with a succession of many generations of houses, apparently without any important break between the IIB and the Ivb stages. It enables us to test the spatial organization models developed by different schools and to propose an alternative model that, according to us, gives a more satisfactory view of the presently available Alsatian data. The two main prevailing models in competition today, and whose validity we question, are the Hofplatz model, which dominated research for a long time 6, and the more recent model of the Zeilensiedlung 7. At the basis of the Hofplatz model, there is the fundamental notion of the Hof 8, which concerns the house itself and a series of associated pits. Among these structures there are the construction pits, an almost log- Archäologisches Korrespondenzblatt 48 · 2018 311 ical fact, but also some other pits which theoretically occupy specific spaces around the building; these are respectively named »northern pits«, »eastern pits« and »western pits«. The aggregation of these scattered pits with the house, at first sight surprising, is simply the result of a wish to create a link between the houses themselves and structures relatively rich in material and thus potentially datable. Actually, one of the main characteristics of the Aldenhoven plateau houses is that they were accompanied by construction pits containing only a very small amount of material. In order to establish an organic link between the »free-standing« pits and the houses, the researchers proposed debatable matchings or pairings since they mainly relied on fabric analysis or on other technical characteristics of the pottery. The links between structures are very rarely illustrated by combinations of evidence. At Langweiler 8, as O. Rück, the principal critic of the model, puts it: the ideal model of the house surrounded by its satellite pits could only possibly be suitable for less than 10 % of the houses 9. The attributions of the pits to such or such a house are considered arbitrary by this author and at least on this point, we will follow him. Relying on these rather weak criteria, the team of the Aldenhoven plateau proposed a definition of the Hof, a concept at the heart of all the developments leading to the Hofplatz system. These researchers defined a yard around the house, corresponding to the space used for farm activities, approx. 25 m in radius. According to that system, two houses separated by fewer than 50 m cannot be considered contemporary. more recently, J. Lüning 10 slightly qualified this view, stating that the Hof area could reach an extension between 50 and 70 m around the house and that its size was likely to change depending on the type of houses defined by P. J. R. modderman 11; thus, the Hof area would be proportional to the size of the building. The successive farms rebuilt generation after generation (every 25 years or so) by the same social segment would move within a definite space: this area would vary between 40 m × 50 m and 80 m × 130 m. On the Langweiler sites, the distances between the successive buildings of a single Hofplatz range between 18 and 60 m. These Hofplätze can develop their own pottery traditions and also show signs of specialisation and significant differences. For a long time, this model could not be ignored and its importance has been reasserted recently 12, but it had to put up with a few but serious methodological criticisms aimed at the legitimacy of the use of material coming from »free-standing« pits to establish the chronology 13. These highly justified criticisms cannot be ignored. The very definition of the Hof as a building surrounded by a 25 m-radius activity area, based on the layout of those same »free pits« can hardly be accepted without debate. In short, as the chronology fails, the whole general model has to be reconsidered, and it is possible that the Langweiler Wohnplätze plans published a great many times may not reflect any past reality simply because the houses were not properly dated. This drastic questioning of the model was initiated by B. Birkenhagen and then taken over by O. Rück. The former only exposes the incoherence and gaps in the Hofplatz model, while the latter develops a new model, the »row model« (Zeilensiedlungsmodell), which due to its own methodical approach gives rise in turn to as much criticism as the model he claims to replace. This most recent counter-model is inclined to favour a »row model« organization for LBK houses, thus breaking spectacularly with the hypothesis of a cycle of regular abandonment and reconstruction 14. O. Rück sets out the proposition that LBK houses can last more than 25 years, perhaps up to 100 years or even longer. In order to build his model, he has to free himself from any chronological constraint, which leads him to this succinct formulation that denies any value to the material assemblages coming from lateral pits: »it is impossible to definitely determine the starting point of a house, let alone its end, on the basis of the pottery from pit fills« 15. Of course, filling the construction pits could only take a few years and it is possible that these structures – even if they remained open for several decades, the upper layers having been removed by erosion – could contain material only relating to the beginning of the building history. We can 312 ph. lefranc · a. denaire · A New Model for the Internal Organization of LBK Settlements admit this, and as we will see in the case of Bischoffsheim, nothing can seriously be against the fact that a house built during the middle LBK can still be occupied at the beginning of the next stage; this is, for instance, Th. Link’s point of view in a paper trying to reconcile Hofplatz and Zeilensiedlungen models 16. O. Rück, however, goes much further by ignoring the chronological data completely: the existence of house rows being enough to establish their contemporaneity. Putting aside all chronological considerations, he builds a non-refutable model. We do not question the existence of house rows known for a long time in the academic literature; neither do we question the assumption according to which such contemporary house rows can have existed, but we underline the extreme fragility of such a model built on methodological arguments at least as questionable as the ones founding the Hofplatz and rightly criticized by O. Rück. a new Model of spatial oRGanization: the »oRthoGonal Model« The Hofplatz asserts the principle of permanence of occupation of the same space and of a long duration of the same social segment (the household). The idea of the reproduction of the same social unit within a defined space is a central idea implying strong stability of settlement and allowing little room for uncertainty. The spatial analysis of the Bischoffsheim site leads us to confirm this principle as valid but also ends up revealing the existence of an original settlement model which we propose naming as »orthogonal model« (figs 6-7). This is based mainly on the chronological data often misused and sometimes ignored in the existing models. The first point to clarify before launching into the construction of a theoretical model of spatial organization concerns the longevity of the houses; on purely material considerations and with a minimum amount of maintenance, that longevity can theoretically reach over a century. The material found in the lateral pits, often not very deep and much eroded, can only give us information about the stylistic stage of the house when it was built. The questions relating to the rhythm of the fillings up and the partial recording of the duration of the house occupation must, of course, be kept in mind. As we know from the results of the Bayesian modelling of the Lower Alsace LBK, about ten generations followed one another between the early LBK (stage IIB) and the end of the late LBK (stage Ivb), ranging between 5320 and 5040 BC: some 280 years. Let us note that the shortest stylistic stages (IIC, Iva1, Iva2 and probably III) can only be seen in Bischoffsheim in a single construction phase; we can thus rightfully estimate that these houses were only occupied by one generation. moreover, in line with its real period (two or three generations), the IIB stage is represented by two, probably three generations of houses; if we count only one phase of construction for the Ivb stage, it is simply because the site was very probably abandoned at the beginning of that stage. The Ivb pottery assemblages are rare and show no sign presaging the final LBK. Combining the duration of each stylistic step with the number of construction phases identified in these same stylistic stages, we can consider that the longevity of a building does not exceed a quarter of a century. This is the theoretical duration of a generation and it is very likely ruled by social codes 17. After referring to the chronological classification of the houses established by the factorial correspondence analysis and having taken into account the total duration of each stylistic stage, we were then in a position to draw up the village map during three short stages (IIC, III and va1). Our first result was that the houses considered by the correspondence analysis as potentially contemporary and which belong to a given stylistic stage, having lasted for one (IIC, Iva1) or maybe two generations (III), are sometimes very close. During the IIC stage, houses 10, 11 and 12 are fewer than 20 m away from each other and during the Ival stage, the average distance between the Großbau type houses is fewer than 30 m, which makes the application of the Hofplatz model extremely problematic. Archäologisches Korrespondenzblatt 48 · 2018 313 fig. 6 The evolution of the Bischoffsheim settlement (dép. Bas-Rhin / F) between stages IIB and IIC. – (Illustration Ph. Lefranc). 314 ph. lefranc · a. denaire · A New Model for the Internal Organization of LBK Settlements fig. 7 The evolution of the Bischoffsheim settlement (dép. Bas-Rhin / F) between stages III and Ivb. – (Illustration Ph. Lefranc). Archäologisches Korrespondenzblatt 48 · 2018 315 The picture we got of the middle LBK stage led us to conceive a model of space partition looking like plots of a quadrangular plan. We can, in fact, notice an alignment of three potentially contemporary houses (houses 3, 33 and 36), very close to each other (only about 10 m separate them), but they are laid out in such a manner that the front walls always stand back from the rear wall of the building ahead. On the other hand, the distance between the long sides of the Großbau type houses is important: about 50 m. near houses 33 and 36, two sorts of Bau type houses can be seen; they are also attributable to the middle LBK. We thought it was justified to consider, as a hypothesis, the possibility that all these elements were contemporary and to divide the space according to a series of parallel lines, about 50 m apart, positioned southwest, northeast and roughly following the lines defined by the front walls and the rear walls of the buildings. moreover, in accordance with the principle of the reproduction of the household at every new generation, we checked if the houses belonging to the following stylistic stages also fitted in with the defined lines. And indeed, the houses attributed to the Iva1 stage and then to the Iva2 / IvB stages are distributed quite harmoniously on both sides of these lines. The closeness between houses 9 and 13, both dated back to the Iva1 stage, enabled us to define a northwest-southeast line dividing the settlement space under consideration into five plots of a quadrangular plan. The first part of the early stage (IIB) shows the remnants of four buildings with enough material to be dated precisely. Eight other houses are probably attributable to stage IIB, but given the poverty of the decorated pottery, an attribution to stage IIC cannot be theoretically completely excluded. It appears obvious that the early LBK houses are not distributed inside the plots we propose; they readily intersect the limits and may be thought a weakness in our interpretation. But if we consider only the early LBK houses, we can notice that the same orthogonal system is at work, but that it organizes itself according to a grid whose orientation is identical to that of the houses; and we know that the change from the early stage to the middle stage is characterized on the site by an important sliding of the orientations towards the east-west line. The southwest-northeast lines dividing the space into strips about 50 m wide have also been defined by a few recognizable alignments (rear gables of houses 32 and 38, of houses 2 and 4, 26 and 27). The southwest-northeast line has been placed according to the distribution of the Großbau houses of stage IIC, a short one, retaining as an axiom the existence of a single house of the Großbau type on every plot. Consequently, we have to consider an initial partition of space during the early stage and then a reorganization of the village at the beginning of the middle one. According to our model, the excavated part of the Bischoffsheim village can be divided into plots of a quadrangular plan, about 50 m × 80 m. Each plot contains one or several buildings. When several buildings are associated there is only one Großbau type building, together with Bau and / or Kleinbau buildings. The houses are occupied for about one generation, and then left and rebuilt within the limits of the plot. The houses can be anywhere in the plot (fig. 8); two houses, contemporary and of the same type, can thus be very close. Therefore, a front wall may only be a few metres away from a rear wall, but always jutting out from one side or the other. On the other hand, the contemporary Großbau type houses are separated by at least 25 m on the southeast-northwest line, that is to say between their gutter or long walls. Even if the case of two houses located at the contiguous ends of two plots, with gutter walls near to each other, has not been observed, it remains theoretically possible. The activity space is not organized all around the house as is the case in the Hofplatz model, but according to its situation in the plot on either side of the gutter walls or on one side of the house only. We cannot avoid reference here to house Iv in Rötha (Lkr. Leipzig / D) surrounded by two enclosed areas developing on both sides of the gutter walls 18 and seeming to materialize in an almost perfect manner one of the plots the existence of which is suspected by us at Bischoffsheim (fig. 9). If only some of the most deeply established enclosed areas are kept, let us bet that this type of development was much more usual 316 ph. lefranc · a. denaire · A New Model for the Internal Organization of LBK Settlements fig. 8 Schematic representation of the »orthogonal model«. – (Illustration Ph. Lefranc). if not omnipresent on LBK settlements. We should then restore the stockade or hedges separating these enclosed areas but also the gardens and the other activity areas located on both sides of these farmhouses, just as we restore the walls or the roofs of the latter. This model is close to the Hofplatz in its principle since it retains the idea of the repetition by translation of the same social unit within a definite space. It also borrows from a Dutch model 19, in acknowledging that each plot can have several buildings, but only of different types; we can rather easily envisage a dependency link between the Großbau type houses on the one hand and the Bau and Kleinbau type houses which may be sheltering other branches of, say, a given lineage, on the other. The »orthogonal model« is less uncertain than the Hofplatz model and seems to be in better agreement with the principle of orthogonality 20: the »daily geometry« 21 which, in pottery decoration, the plans of the houses, the layout of the graves or the structure of the enclosures, characterizes the LBK. To our mind, the »orthogonal model« has another advantage: simplicity. It accounts, setting aside the idea of some proto-urbanism, for the alignments of buildings observed at Sierentz (dép. haut-Rhin / F), Entzheim (dép. Bas-Rhin / F), Archäologisches Korrespondenzblatt 48 · 2018 317 fig. 9 Rötha (Lkr. Leipzig / D), house Iv and its stockades. – (After Dalidowski et al. 2016, 72 fig. 12). Langweiler 2, Langweiler 9, Straubing (Lkr. Straubing-Bogen / D), Elsloo (prov. Limburg / nL) and elsewhere. These alignments are merely due to constraints as the household can only reproduce itself architecturally inside a well-defined space; thus, the houses are rather logically rebuilt next to each other, thus sometimes forming those village-street plans which have no reality and the origins of which have already been well explained by J. Lüning. Following our model, it is possible to outline the history of the village of Bischoffsheim, founded around 5300 BC. houses 34 and 16 could be pioneer houses at the root of the houses situated on plots 3 and 5. These houses, set in the same northwest-southeast direction, are the only ones with an orientation between 283 and 293 °, an orientation also known in Alsace in Achenheim (dép. Bas-Rhin / F) and Rosheim (dép. BasRhin / F) and concerning houses equipped with an Außengraben, a typical earlier LBK feature. Basing ourselves on the Bayesian chronological model, we accept the hypothesis of a relatively long IIB stage (2 generations after the pioneer generation) and thus propose considering the houses of the same type localised on the same plots as diachronic and successive. Their succession over time can, however, not be defined. So, we propose an interpretation of the plan which can only be a hypothesis. We stress, however, the fact that over all the settlement duration, plot 3 exclusively holds houses of the Bau type and that plot 5 shows two, perhaps three, repetitions of the pair Großbau + Bau or Kleinbau; that would represent a good exam- 318 ph. lefranc · a. denaire · A New Model for the Internal Organization of LBK Settlements ple of distribution by translation of the same social unit. In stage IIC, plot 3 may no longer be occupied (unless Kleinbau 10, which could be linked with Großbau 12 of plot 5, has followed Bauten 34, 32 and 38). The other plots all hold Großbau type houses including some very big ones. At the beginning of the middle stage, the plots are reorganized according to the adopted planning. Three of the studied plots (1, 3 and 5) show connected houses under the rule Großbau + Bau 22. During stage Iva1 which covers a generation, the connected houses are only found on plot 3 and plots 2, 4 and 5 hold Großbauten. Plot 1 is definitely abandoned. The houses of stage Iva1 appear to be an assemblage of rather clustered buildings, the average distance between contemporary houses of type Großbau being c. 25 m. Phase Iva2 is represented on plots 3 and 4, where houses 14 and 28 respectively replace houses 9 and 22. At that time, plot 5 is still occupied by house 8 which will be rebuilt later, at the beginning of stage Ivb (house 31). It is highly probable that houses 14 and 28, rebuilt at stage Iva2 (duration: c. 10 years), are still lived in. The number of contemporary houses in the excavated area can be estimated between two (at the foundation), five or six (in the early and middle LBK) and four (in the late LBK). Whatever the calculation methods 23, and if we privilege the central part as the main living space, the number of inhabitants can be calculated, depending on construction phases, as 50-80 persons. As only half of the site has been explored it can be estimated that it was a village with 100-160 persons. All these results, even if they are imprecise and hypothetical, are in accordance with the image of small communities shown by the necropolis including the largest one used during the recent stage, phase Iva1 to phase Ivb and which contains a hundred people 24. A total population of about 150 individuals is also in accordance with the ethnographic data as the population of traditional villages rarely exceeds 250 inhabitants. The alternative hypothesis proposed here has, like any other model, its strong and weak points. We think that it can account for the configurations observed at Bischoffsheim, but also on two other sites in Alsace excavated on a large scale which have produced well-dated houses: at Entzheim »Les Terres de la Chapelle« 25 and at Sierentz »Sandgrube« 26. We do not assert a universal validity for the model right away, but the application of the model may well give good results on the Aldenhoven plateau and elsewhere. In order to check this, it would be necessary to work again on all the chronological data concerning classic sites such as Langweiler 8 or Elsloo – but long and relatively difficult research would have to be conducted. Finally, the model set out here does not go against the theory inspired by »exogamous moieties« and developed among others by h.-Ch. Strien 27; it is also not completely incompatible with the »wards model« (modèles des quartiers / Quar­ tiersmodell) of P. van the velde, but both these interpretations are on a different level of analysis. acknowledgements The authors thank for the help of Jocelyne Lefranc and Alasdair Whittle (Cardiff). notes 1) Lefranc 2007a. 6) Boelicke 1988. – neth 1999. – herren 2005. – Classen 2005. 2) Lefranc 2007b. 7) Rück 2013. 3) meier-Arendt 1966. 8) Boelicke et al. 1988. – Boelicke 1988. 4) Lefranc 2007a. 9) Rück 2013. 5) Denaire et al. 2017. 10) Lüning 1998. Archäologisches Korrespondenzblatt 48 · 2018 319 11) modderman 1970. 15) Rück 2013, 210. 22) This hypothesis is conceivable only if the middle stage lasts a little more than a generation or if there is a hiatus in the occupation of the site. If the middle stage includes two generations and the occupation is continuous, one has to imagine that certain plots at some time or other, only hold Bau type houses. In that case, too, the possible succession of the houses cannot be determined. 16) Link 2012. 23) Coudart 1998. – Dubouloz 2008. – Bocquet-Appel et al. 2014. 12) Lüning 2005; 2012. – Classen 2005. 13) Birkenhagen 2003, 50. 14) Rück 2007; 2013. 17) Descola 1986. 18) Dalidowski et al. 2016, 72 fig. 12. 19) van de velde 1990. 24) Boës et al. 2007. 25) Lefranc et al. 2017. 20) Jeunesse 2009. 26) Lefranc 2015. 21) van Berg / Cauwe 1996. 27) Strien 2005. References van Berg / Cauwe 1996: P.-L. van Berg / n. Cauwe, Du néolithique ancien au néolithique récent en Europe occidentale: changement de géométrie. In: P. Duhamel (ed.), La Bourgogne entre les bassins rhénan, rhodanien et parisien: carrefour ou frontière? Actes du 18e colloque interrégional sur le néolithique, Dijon, 25-27 octobre 1991. Revue Archéologie de l’Est et du Centre-Est Supplément 14 (Dijon 1996) 55-78. Birkenhagen 2003: B. Birkenhagen, Studien zum Siedlungswesen der westlichen Linearbandkeramik. Saarbrücker Beiträge zur Altertumskunde 75 (Bonn 2003). Bocquet-Appel et al. 2014: J.-P. Bocquet-Appel / J. Dubouloz / R. moussa / J.-F. Berger / A. Tresset / E. Ortu / J.-D. vigne / R. Bendrey / S. Bréhard / D. Schwartz / A. Salavert / m. F. Sanchez-Goñi / D. Ertlen / Y. Gauvry / G. Davtian / m. vander Linden / E. Lenneis / L. noiret / A. Guillaumont / m. O’Connor, multi-agent modelling of the Trajectory of the LBK neolithic: A Study in Progress. In: A. Whittle / P. Bickle (eds), Early Farmers: The view from Archaeology and Science. Proceedings of the British Academy 198 (Oxford 2014) 53-69. Boelicke 1988: u. Boelicke, Die Gruben. In: Boelicke et al. 1988, 300-394. Boelicke et al. 1988: u. Boelicke / D. von Brandt / J. Lüning / P. Stehli / A. Zimmermann, Der bandkeramische Siedlungsplatz Langweiler 8, Gemeinde Aldenhoven, Kreis Düren. Rheinische Ausgrabungen 28 = Beiträge zur neolithischen Besiedlung der Aldenhovener Platte 3 (Köln, Bonn 1988). Boës et al. 2007: E. Boës / C. Jeunesse / R.-m. Arbogast / Ph. Lefranc / m. mauvilly / F. Schneikert / I. Sidéra, vendenheim, »Le haut du Coteau« (Bas-Rhin): remarques sur l’organisation interne d’une nécropole du néolithique ancien danubien. In: m. Besse (ed.), Actes du 27e colloque interrégional sur le néolithique, neuchâtel 2005. Cahiers d’Archéologie Romande 108 (Lausanne 2007) 279-283. Classen 2005: E. Classen, Siedlungsstrukturen der Bandkeramik im Rheinland. In: Lüning / Fridrich / Zimmermann 2005, 113-124. Coudart 1998: A. Coudart, Architecture et société néolithique: l’unité et la variance de la maison danubienne. Documents d’Archéologie Française 67 (Paris 1998). 320 Dalidowski et al. 2016: m. Dalidowski / A. homann / Th. Laurat / h. Stäuble / C. Tinapp, Linienbandkeramische häuser bei hain und Rötha, Lkr. Leipzig. Ausgrabungen in Sachsen 5, 2016, 62-77. Denaire et al. 2017: A. Denaire / Ph. Lefranc / J. Wahl / Ch. Bronk Ramsey / E. Dunbar / T. Goslar / A. Bayliss / n. Beavan / P. Bickle / A. Whittle, The cultural project: formal chronological modelling of the Early and middle neolithic sequence in Lower Alsace. Journal of Archaeological method and Theory 24, 2017, 10721149. Descola 1986: P. Descola, La nature domestique. Symbolisme et praxis dans l’écologie des Achuar (Paris 1986). Dubouloz 2008: G. Dubouloz, Impacts of the neolithic Demographic Transition on Linear Pottery Culture Settlement. In: O. Bar Yosef / J.-P. Bocquet-Appel (eds), Conference at the harvard university Center for the Environment, harvard 2006 (new York 2008) 207-235. herren 2005: B. herren, harting-nord, Kr. Regensburg / Oberpfalz. Keine Krise am Ende der Linearbandkeramik! Kontinuität und Diskontinuität am Übergang von Alt- zu mittelneolithikum. In: Lüning / Fridrich / Zimmermann 2005, suppl. 5. Jeunesse 2009: C. Jeunesse, Le front de colonisation occidental (entre Rhin et Seine) et l’identité rubanée. Réflexion sur les systèmes symboliques dans le néolithique danubien. In: J. Kozlowski (ed.), Interactions between different models of neolithization north of the central european agro-ecological barrier. Symposium Kraków 2008. Prace Komisji Prehistorii Karpat 5 (Kraków 2009) 151-176. Lefranc 2007a: Ph. Lefranc, L’habitat rubané de Bischoffsheim (BasRhin), premiers résultats. In: F. Le Brun-Ricalens / F. valotteau / A. hauzer (eds), Relations interrégionales au néolithique entre Bassin parisien et Bassin rhénan. Actes du 26e colloque interrégional sur le néolithique, Luxembourg 2003. Archaeologia mosellana 7 (Luxembourg 2007) 9-21. 2007b: Ph. Lefranc, La céramique du Rubané en Alsace: contribution à l’étude des groupes régionaux du néolithique ancien dans la plaine du Rhin supérieur. monographies d’Archéologie du Grand Est 2 (Strasbourg 2007). ph. lefranc · a. denaire · A New Model for the Internal Organization of LBK Settlements 2015: Ph. Lefranc, Les villages du néolithique ancien en Alsace: un état de la recherche. Archéopages 40, 2015, 18-25. Lefranc et al. 2017: Ph. Lefranc / F. Chenal / A. Denaire / E. Guthmann / n. Schneider, L’habitat et les sépultures du néolithique ancien Rubané d’Entzheim »Les Terres de la Chapelle« (BasRhin). Revue Archéologique de l’Est 66, 2017, 17-74. Link 2012: Th. Link, »hofplatz« und »Zeilensiedlung«: konkurrierende modelle oder zwei Seiten derselben medaille? In: F. Kreienbrink / m. Cladders / h. Stäuble / T. Tischendorf / S. Wolfram (eds), Siedlungsstruktur und Kulturwandel in der Bandkeramik. Beiträge der internationalen Tagung »neue Fragen zur Bandkeramik oder alles beim Alten?!«; Leipzig, 23. bis 24. September 2010. Arbeits- und Forschungsberichte zur sächsischen Bodendenkmalpflege Beihefte 25 (Dresden 2012) 43-46. Lüning 1998: J. Lüning, L’organisation régionale des habitats rubanés: sites centraux et sites secondaires (groupements de sites). In: n. Cauwe / P.-L. van Berg (eds), Organisation néolithique de l’espace en Europe du nord-Ouest. Actes du 23e colloque interrégional sur le néolithique; Bruxelles, 24-26 octobre 1997. Anthropologie et Préhistoire 109 (Bruxelles 1998) 163185. 2005: J. Lüning, Bandkeramische hofplätze und die absolute Chronologie der Bandkeramik. In: Lüning / Fridrich / Zimmermann 2005, 49-74. 2012: J. Lüning, Bandkeramische hofplätze und Erbregeln. In: T. L. Kienlin / A. Zimmermann (eds), Beyond Elites – Alternatives to hierarchical Systems in modelling Social Formations. International Conference at the Ruhr-universität Bochum, Germany, October 22-24, 2009. universitätsforschungen zur Prähistorischen Archäologie 215 (Bonn 2012) 197-201. Lüning / Fridrich / Zimmermann 2005: J. Lüning / Ch. Fridrich / A. Zimmermann (eds), Die Bandkeramik im 21. Jahrhundert: Symposium in der Abtei Brauweiler bei Köln vom 16.9.19.9.2002. Internationale Archäologie. Arbeitsgemeinschaft, Symposium, Tagung, Kongress 7 (Rahden / Westf. 2005). meier-Arendt 1966: W. meier-Arendt, Die bandkeramische Kultur im untermaingebiet. veröffentlichungen des Amtes für Bodendenkmalpflege im Regierungsbezirk Darmstadt 3 (Bonn 1966). modderman 1970: P. J. R. modderman, Linearbandkeramik aus Elsloo und Stein. Analecta Praehistorica Leidensia 3 (Leiden 1970). neth 1999: A. neth, Eine Siedlung der frühen Bandkeramik in Gerlingen, Kreis Ludwigsburg. Forschungen und Berichte zur vorund Frühgeschichte in Baden-Württemberg 79 (Stuttgart 1999). Rück 2007: O. Rück, neue Aspekte und modelle in der Siedlungsforschung zur Bandkeramik. Die Siedlung Weisweiler 111 auf der Aldenhovener Platte, Kr. Düren. Internationale Archäologie. Arbeitsgemeinschaft, Symposium, Tagung, Kongress 105 (Rahden / Westf. 2007). 2013: O. Rück, From yard to house row: the Bandkeramik village – layouts in rows and feature-free areas provide a new view on settlement structure. In: C. hamon / P. Allard / m. Ilett (eds), The Domestic Space in LBK Settlements. Internationale Archäologie. Arbeitsgemeinschaft, Symposium, Tagung, Kongress 17 (Rahden / Westf. 2013) 201-230. Strien 2005: h.-Ch. Strien, Familientraditionen in der bandkeramischen Siedlung bei vaihingen/Enz. In: Lüning / Fridrich / Zimmermann 2005, 189-197. van de velde 1990: P. van de velde, Bandkeramik social inequality – a case study. Germania 68/1, 1990,19-38. Zusammenfassung / Summary / Résumé Ein neues Modell für die interne Organisation von LBK-Siedlungen: der Fundplatz von Bischoffsheim (dép. Bas-Rhin / F) und das »orthogonale Modell« Die räumliche Analyse des neolithischen Dorfes von Bischoffsheim führte zur Entwicklung eines neuen modells interner Organisation von Siedlungen der Linearbandkeramischen Kultur, das auf einer Serie exakt datierter häuser basiert sowie auf einem Bayesschen modell zahlreicher Radiokarbondatierungen. Das »orthogonale modell« – im Prinzip ähnlich zum hofplatz-modell – spiegelt auch die Reihung wider, die an vielen Plätzen beobachtet wurde. Es postuliert eine regelmäßige unterteilung des Dorfareals in viereckige Grundstücke mit Abfolgen von Gebäuden, die regelmäßig (ungefähr jedes vierteljahrhundert) durch dieselbe soziale Einheit neu erbaut wurden. Übersetzung: m. Struck A New Model for the Internal Organisation of LBK Settlements: the Site of Bischoffsheim (dép. Bas-Rhin / F) and the »Orthogonal Model« The spatial analysis of the neolithic village of Bischoffsheim led to the development of a new model of internal organisation of the Linear Pottery culture settlements based on a series of precisely dated houses and on a Bayesian modelling of numerous radiocarbon dates. The »orthogonal model«, similar in principle to the Hofplatz model but also reflecting the row organisation observed at many sites, proposes the partition of the village, which displays a regular division into quadrangular lots with successive buildings. These were rebuilt repeatedly (approximately every quarter of a century) by the same social unit. Archäologisches Korrespondenzblatt 48 · 2018 321 Un nouveau modèle d’organisation interne des habitats du Rubané: le site de Bischoffsheim (dép. Bas-Rhin / F) et le »modèle orthogonal« L’analyse spatiale de l’habitat néolithique ancien de Bischoffsheim a débouché sur l’élaboration d’un nouveau modèle d’organisation interne des habitats rubanés reposant sur une série de bâtiments précisément datés et sur une analyse bayésienne de nombreuses dates radiocarbone. Le »modèle orthogonal«, proche dans son principe du modèle du Hof­ platz mais permettant également de rendre compte de l’organisation en rangées observée sur de nombreux sites, propose de reconnaître une partition régulière de l’espace villageois divisé en lots quadrangulaires accueillant des bâtiments successifs, régulièrement reconstruits (tous les quarts de siècle environ) par la même unité sociale. Schlüsselwörter / Keywords / Mots clés Frankreich / Elsass / neolithikum / Linienbandkeramik / Siedlungsorganisation / LBK-Chronologie / Bayessches modell France / Alsace / neolithic / Linear Pottery culture / settlement organisation / LBK chronology / Bayesian modelling France / Alsace / néolithique / céramique linéaire / organisation des habitats / chronologie / modélisation bayésienne Philippe Lefranc Anthony Denaire Institut national de recherches archéologiques préventives Grand-Est nord, région Alsace Centre archéologique de Strasbourg et umR 7044 Archimède, université de Strasbourg 10, rue d’Altkirch F - 67100 Strasbourg philippe.lefranc@inrap.fr université de Bourgogne umR 6298 ArTehiS 6, boulevard Gabriel F - 21000 Dijon anthony.denaire@u-bourgogne.fr 322 ph. lefranc · a. denaire · A New Model for the Internal Organization of LBK Settlements INHALTSVERZEICHNIS Ivor Janković, Nikola Vukosavljević, James C. M. Ahern, Ivor Karavanić, Sanjin Mihelić, Fred H. Smith, Bukovac Cave Revisited: Recent Excavations of an Early Upper Paleolithic Site in the Gorski Kotar Region of Croatia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297 Philippe Lefranc, Anthony Denaire, A New Model for the Internal Organization of LBK Settlements: the Site of Bischoffsheim (dép. Bas-Rhin / F) and the »Orthogonal Model« . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307 David Parma, Peter Barta, Ivana Jarošová, Sylva Kaupová, Miriam Nývltová Fišáková, Lenka Vargová, A Cruel Fate? The Unique Bronze Age Burial from Ivanovice na Hané (okr. Vyškov / CZ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323 Dagmara Łaciak, Früheisenzeitliche Gefäße vom Typ kernos: Versuch einer funktionalen Interpretation aufgrund eines Neufunds in Domasław (woj. dolnośląskie / PL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339 Przemysław Dulęba, Magdalena Konczewska, Paweł Konczewski, Jacek Szczurowski, Agnieszka Tomaszewska, A Przeworsk Culture Inhumation Burial from Siechnice (Lower Silesia) and the Question of Biritualism in the Pre-Roman Iron Age . . . . . 351 Carsten Mischka, Alexander Rubel, Rada Varga, Das Lager der ala I Batavorum milliaria und sein vicus in Războieni-Cetate (jud. Alba / RO). Geophysikalische Untersuchungen und historische Einordnungen . . . . . . . . . . . . . . . . . . . . . . . 377 Ferdinand Heimerl, Bernd Päffgen, Zwei Neufunde frühmittelalterlicher Glaskameen aus Trier und Bitburg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407 Zbigniew Robak, Two Carolingian Strap-ends on Exhibition in Želiezovce (okr. Levice / SK) . . . . . . . . . 417 Elisabeth Nowotny, Ein frühmittelalterlicher Depotfund vom Schanzberg von Thunau (Bez. Horn / A) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437 ISSN 0342-734X