High Resolution AMS Dates from Shubayqa 1, northeast Jordan Reveal Complex Origins of Late Epipalaeolithic Natufian in the Levant

The Late Epipalaeolithic Natufian (~14,600 − 11,500 cal BP) is a key period in the prehistory of southwest Asia. Often described as a complex hunting and gathering society with increased sedentism, intensive plant exploitation and associated with an increase in artistic and symbolic material culture, it is positioned between the earlier Upper- and Epi-Palaeolithic and the early Neolithic, when plant cultivation and subsequently animal domestication began. The Natufian has thus often been seen as a necessary pre-adaptation for the emergence of Neolithic economies in southwest Asia. Previous work has pointed to the Mediterranean woodland zone of the southern Levant as the ‘core zone’ of the Early Natufian. Here we present a new sequence of 27 AMS radiocarbon dates from the Natufian site Shubayqa 1 in northeast Jordan. The results suggest that the site was occupied intermittently between ~14,600 − 12,000 cal BP. The dates indicate the Natufian emerged just as early in eastern Jordan as it did in the Mediterranean woodland zone. This suggests that the origins and development of the Natufian were not tied to the ecological conditions of the Mediterranean woodlands, and that the evolution of this hunting and gathering society was more complex and heterogeneous than previously thought.


Methods
Excavations at Shubayqa 1 were carried out using a 1 x 1 meter grid system to ensure a high accuracy in recording the spatial distribution of finds. Sediments were excavated using a combined system of singe-context recording and excavation in arbitrary levels in each square meter. Thus, thick deposits were sub-divided into smaller excavation spits, but natural boundaries of sediments were always respected. Spoil from the excavations was dry-sieved on site using a 1 x 1 mm mesh. Sediment samples were collected from all excavated contexts for flotation, phytolith and pollen analysis. Wherever possible samples for charred plant remains for 14 C dating were collected in the field and placed into sterile sample containers. Alternatively, they were taken from bulk botanical samples in those cases were samples could not be collected on-site from some contexts or where those samples proved unsuitable for dating. The provenience of every sample collected during actual excavation was recorded noting their position on the site grid and elevation above sea level.
Where possible charred plant remains were identified to genus or species level. Even where this was not possible only short-lived parts of plants (seeds or twigs) were used for dating to avoid use of old Dates were calibrated in OxCal 4.3 using the IntCal13 calibration curve. We provide dates at the 1σ range of 68.2% probability throughout the text. For the 2σ range please see the supplementary files.
Using the position of each date in the stratigraphic matrix of the site, we were able to further constrain the dates using the phase function in OxCal 4.3 to create a Bayesian model for Shubayqa 1. To examine changes in lithic typology we recorded all complete and broken lunates from all the phases represented in the main stratigraphic sequence at the site. All lunates were categorized as either with or without Helwan retouch. Length and width measurements were taken using calipers on all complete lunates.

Phase Description
Radiocarbon Dates 7 earliest phase; associated with the construction and use of Structure 1. Finds related to the occupation and use of Structure 1 were recovered from the building's stone-lined fireplace, as well as the construction cut of the structure's exterior wall. The two fills of the fireplace were particularly dense in material culture, faunal remains and especially charred plant remains.

RTD-7951
Beta-112146 RTD-7317 RTD-7318 RTD-7948 6 begins after the abandonment of Structure 1. A thin layer of aeolian sediment accumulated above the floor of Structure 1 along the western wall. A thick clayey silt deposit accumulated above this aeolian layer, which contained few finds. Although finds from this deposit were rare, it appears to have an organic and anthropogenic character. This suggests that occupation at the site may have shifted laterally to another part of the site, but that the site continued to be used gradual burial of the Phase 5 fireplace and trampled surface by the accumulation of a c. 0.4 m thick silty clay deposit above the fireplace, which has a low to average finds density. Similar to Phase 6, it is suggested that occupation at the site continued, but shifted to another part of the site.

RTK-6813 RTK-6816
3 begins with the construction of Structure 2 and includes its use. It ends with the abandonment of the building. The boundary between Phase 2 and 3abandonment of structure and start of midden formation -is indistinct and difficult to separate. During this phase the pavement slabs of Structure 2 were repeatedly lifted and the remains of eleven neonates and infants were interred beneath and on top of the floor.

RTK-6819
2 consists of a midden deposit which yielded large assemblages of material culture and fauna. The midden overlay the flagstone pavement of Structure 2. Limited post-Epipalaeolithic material culture occurs in the upper-most part of the deposit, suggesting some recent disturbances; however, they are absent from the lowest layers of the midden. In addition to chipped stone and faunal remains, many ground stone tools were found lying on top of the pavement of Structure 2, although it is unclear whether they were left in situ after Structure 2 fell out of use or whether they accumulated inside the structure as part of waste disposal and general midden accumulation. Disarticulated human remains were also found lying on top of the pavement within the midden.
RTK-6812 RTK-6817 1 corresponds to the occupation exposed in Area C. In this small 1x2m sounding a charcoal rich deposit was located that produced a limited number of artefacts and other finds, and may represent an ephemeral occupation of the site. However, exposure to date has been too limited to fully understand the relationship of this deposit to other parts of the site.