Introduction

This research aimed to define the use of some specific, pointed tools found in the Bronze Age levels of the site of Aradetis Orgora in the Shida Kartli province of Eastern Georgia (Caucasus) (See Figure 1). Five “needles” with different morphological features were unearthed at the site during the recent excavations carried out by the joint GISKAP project of Ca’ Foscari University of Venice in collaboration with the Georgian National Museum (Tbilisi) (Gagoshidze, 2001; 2012; Gagoshidze and Rova, 2015; 2018a; 2018b). In fact, in two of these, the needles proximal part and the eye are more or less in accordance with the common idea of a needle. In the remaining ones, the proximal part has a quadrangular shape and a much larger size. This type of needle is not very common, and objects with similar characteristics are traditionally classified in the literature as pins (Luik, 2013, p.29; Birò, 2005, p.204). The preliminary analysis of functional traces showed, however, a resemblance to the experimental traces of baskets and plant processing in the writer’s experimental comparison collection. Starting from this data it was decided to deepen the study with experimental protocol to verify if that tools had been used as basketry tools.

In order to confirm (or disprove) the hypothesis, two examples of these tools were reproduced: one of the Early Bronze Age (Kura-Araxes period, beginning of the 3rd millennium BC) with a semi-circular proximal part, and one of the Late Bronze Age (second half of the 2nd millennium BC) with a quadrangular proximal part. To corroborate my hypothesis, different techniques of basketry were reproduced with the help of an expert craftsperson.

Thanks to the experimental reproduction of some craft activities it was possible to understand how these tools could be or not be used, confirming the preliminary traceological data.

The archaeological context

The site of Aradetis Orgora lies in the Shida Kartli province of Eastern Georgia and is located at the confluence of the River Kura with the Western Prone, one of its main tributaries. The site, whose first occupation dates to the Kura-Araxes culture (end of the 4th/beginning of the 3rd millennium BC), was the seat of a monumental, fortified palace of the Hellenistic/Early Roman period (See Figure 2a) and continued to be occupied until the Early Middle Age (6th century AD). The ancient settlement developed on three adjacent mounds; the main, Western Mound, also called Dedoplis Gora (the Queen’s Hill) (Gagoshidze, 2001; 2012), the Eastern Mound and the Northern Mound. To the north of the Northern Mound lies the ancient cemetery of Doghlauri. During the excavation campaigns of the Georgian Italian Shida-Kartli Archaeological Project (GISKAP), carried out from 2013 to 2016 (Gagoshidze and Rova 2015; 2018a, 2018b) on the Western Mound, two different areas were excavated: Field A (with three squares of 5 x 5 m oriented in NS direction) on the North-West side of the mound, and Field B (with 4 squares of 5 × 5 m oriented in WE direction) on its eastern side. The excavations confirmed that the most important occupation phases of the site correspond to the Kura-Araxes period (second half of the 4th-first half of the 3rd millennium BC), with a 4-meter thick occupational sequence, and the Late Bronze/Early Iron Age Periods (15th-7th centuries BC), which apparently correspond to the main phases of use of the neighbouring cemetery area as well. A less intensive frequentation of the site in the later Early Bronze Age (Bedeni Culture)is also evidenced and, for the first time in Shida Kartli region, in the Middle Bronze Age (first half of the 2nd millennium BC) (Rova, 2016; Gagoshidze and Rova, 2018a, 2018 b; Aquilano, Gavagnin and Gervasi, 2019).

The Kura-Araxes occupation is characterised by wattle-and-daub round-shaped and rectilinear structures of mainly domestic character but includes a small building with a possibly cultic function (See Figure 2b), in which two zoomorphic ritual vessels have been found (See Figure 2c) (Kvavadze, et al., 2019).

During the following periods (especially during the Late Bronze/Early Iron Age periods, 15th-7th centuries BC), the site was remodelled: stone walls were built along its perimeter, which were interpreted as belonging to a terracing system aimed at enlarging the available space on top of the mound. In fact, in this period, the site shows an important increase in population, as suggested by the large number of contemporary graves discovered in the nearby cemetery (Doghlauri). The excavated areas on both sides of the Main Mound brought to light different open areas located at its periphery, where multiple cooking fires devoted to food preparation were found (See figure 2d).

The temporal continuity of the site’s occupation is a clear sign of its importance within the surrounding region. The wealth of the site in the Kura Araxes period well accords with the presence in the neighbouring area of several contemporary sites with similar characteristics. This shows the importance of the Shida Kartli region, and of the River Kura Valley, in the Kura Araxes as well as in later periods as a centre of stable and relatively intense occupation, which took part in a network of far-reaching connections (Rova, 2014; Gagoshidze and Rova, 2018).

The archaeological bone tools

84 objects of hard material of animal origin were found at the site. Only six out of these are needles, of which two from the Early Bronze Ages (Kura-Araxes levels) (2298 M-1; 2730 M-3), three from the Late Bronze (1850 M-2; 2524 M-1; 2648 M-1) and just one dates from the Iron Age (1354 M-2), (See Table 1 and Figure 3).

Table 1. Aradetis Orgora needles and their measurements. 

The needles belonging to the levels of the Late Bronze Age have all been produced by the shaft of the long bones of ruminants of sheep/goat size. They are very similar to each other in morphology and are well distinguishable from the needles belonging to the other phases of life of the site. They have a regular cylindrical body. The quadrangular proximal part is wider than the body, and they are provided with a quite large hole.

The Kura-Araxes needles, also manufactured from Ovis/Capra diaphysis, differ from those of the later period in their lack of a quadrangular proximal part. Only in one case does the proximal part have slightly larger dimensions than the rest of the instrument and a more circular shape (2298 M-1).

Similar tools had been found in a Late Bronze Age site in Latvia. These tools have a circular or semi-circular proximal part with one or two very wide eyes. They have been interpreted as pins (Luik 2013, 29) (See Figure 4a).

Others similar tools, dated to more recent periods, come from Hungary, where the archaeologists found some bronze pointed tools with a wider proximal part and very large holes: these were interpreted as pins as well (Birò, 2005, p.204) (See Figure 4b).

The functional traces found on the surfaces of these Aradetis tools, however, are similar to those found on the surface of a experimental needle used in experimental basketry activity and in general in the processing of plants.

For this reason, my hypothesis is that these specific tools may have been needles that, owing to their special shape, were probably used in basketry. The hypothesis is corroborated by the shape of modern basketry needles, on which the part with the eye is also wider than the rest of the tool. Therefore, it was essential to proceed with an experimental protocol that may confirm or refute the hypothesis.

Methods

Trace analysis

A preliminary study of the surfaces of the Aradetis Orgora needles has been carried out to identify any techno-functional traces, thanks to which it would be possible to better characterise not only the production techniques but also the use of these objects. The surfaces have been analysed with a metallographic microscope (Nikon Y-IDP e Nikon Japan 261608 with a camera: Olympus SC 100 and Nikon Digital DXM 1200) and a digital microscope (Hyrox RH 2000), thanks to which it was possible to identify the traces on their surfaces (Buc 2011, 546).

Experimental protocol

For this study, it was decided to reproduce only two experimental needles: the one of the Kura-Araxes period (2298 M-1) with a semi-circular proximal part and one of the Late Bronze Age ones (2648 M-2) with a quadrangular proximal part.

The experimental protocol was divided into two steps.

The first was the technological chain that has been further divided into two sub-phases (Provenzano 1999):

Débitage: extraction of the coarse support by detaching it from the beam.

Façonnage: the shaping of the support, active edge finishing, and perforation.

The second step was the focus of this study, functional reproduction. For this phase, two craft activities have been carried out: one on leather (to sew it), and the other on vegetal fibres used in basketry.1

Preliminary trace analysis

All the needles from Aradetis Orgora were analysed with a microscope to search for any techno- functional traces. Where this was possible, the analysis highlight techno- and functional traces. The analysis of manufacturing techniques showed abrasion caused by the use of a ground stone during the façonnage phase (See Figure 5a, b, c, d), while it showed that metal tools were used to re-sharpen the tips of the needles and for producing the holes (See Figure 6a, b, c, d, e).

The functional traces showed a close analogy with experimental traces of basket- making.2  The traces on the needles of the Late Bronze Age levels (n. 1850 M-2 and n. 2648 M-2) are located on all the active and mesial areas of the instrument. They are characterised by a flat relief only with a more wavy point and by a rough to smooth polishing, especially on the distal end, while in the mesial area, this seems to be smoother. Long, thin, and shallow linear depressions are scattered over the whole surface and arranged on the transverse axis of the instrument, both isolated and crossed. Non-linear depressions scattered throughout the affected area and composed of holes and micro-holes with homogeneous and regular morphology are well visible together with the striation (See Figure 7a, b). In the case of needle n. 2648 M-2, the edge of the hole also seems affected by a functional trace that could be related to its use on plants (Christidou, 2005, p.97). The edge is very rounded, and the surface has an angular relief with wavier spots. The texture is rough, and only the highest points tend to be smooth. There are striations arranged in bands that overlap each other. The polishing has a medium to low gloss (See Figure 7c).

Needle n. 2298 M-1 from the Early Bronze Age levels shows a slight morphological difference from needles produced during the following period. In fact, the surface has a more angular relief and a much rougher texture than the needles of the later periods. Also, in this case, the brilliance of the polishing is medium to high, with striations concentrated in the medium-distal area of the instrument, arranged along its transverse axis. Non-linear depressions formed by holes and micro-holes with a homogeneous morphology also concentrate at the points where there is a larger presence of striations. In this case, as well, the archaeological trace is very similar to the experimental one related to the handicraft activity of basketry and in general in the processing of plants (See Figure 7d, e, f).

The technological experimentation

The raw material came from different long bones of a cow,3  from which long splinters were obtained. The extraction of the blank was carried out with modern metal saws (See Figure 8a).

During the façonnage phase, two techniques were used: scraping and abrasion. For the first technique, I used a modern metal chisel to eliminate the hardest parts of the blank in order to give it a preliminary circular shape and thus shorten the manufacturing time (See Figure 8b). For the second technique, I used limestone with sand, and the blank was wetted several times in order to increase the power of abrasion needed to shape the object and, specifically, to create the active part (See Figure 8c).

The last step was to create the eye in the proximal part. The technological traces identified in the previous analysis led to the hypothesis that they were created with a drill-type mechanical tool. For this reason, I used a pump drill equipped with a bronze point4  (See Figure 8d).

The replicas obtained were very similar to the archaeological tools.

The functional experimentation

Premise

This step is the focus of this study. Starting from the preliminary analysis of traceological traces each experimental needles was used on two different materials in order to test other variables that could demonstrate their possible function as needles for leatherworking or basketry. For this reason, they were first tried out with the perforation of dry leather and then were used for basketry activities.

Use on leather

The reproduced needles were used for stitching soft and semi-hard leather. The leather employed was of bovine origin and was used both in a dry state and in a softened state. The use of leather in different states allowed a better evaluation of the tools. Both needles were used in activity for about 20 minutes.

During the activity, both needles responded similarly to the processing. In fact, in both cases, the proximal part was too wide. During the drilling process, in which the eye was passed, it was necessary to employ a rotary movement of the needles, which required considerable effort. However, this action, performed either with the square eye needle or with the other, left very big holes in the leather. The author has carried out moulds on the surfaces of the needles used on the leather, with the liquid silicon Provil in such a way as to be able to have a faithful copy of the traces refractive to the processing of the leather.

The difficulties encountered in this activity with different types of leather suggest that both reproductions of the Aradetis Orgora needles were inadequate for drilling leather (See Figure 9a, b, c).

Use in basketry

The contribution of an expert craftsperson in the art of basketry was necessary for this experimental phase. The experimental part was thus performed by artisan Giovanni Morra, who used both needle replicas in the production of some baskets.5  Both needles were used in these activity for about 40 minutes.

Giovanni chose to reproduce two of the main known techniques of basket making:

•  sewed coiling;
•  sewed-braid coiling.

These techniques are quite simple and were also commonly used in ancient times, for instance, during the Neolithic period. They are mainly used with soft vegetables, in particular with aquatic plants (Wendrich, 1991, p.54; Wendrich and Ryan 2012, p.57).

In the absence of precise data on the type of vegetation present in the area surrounding the site during the Bronze Age6 , the artisan used a very common plant Ampelodesmos mauritanicus for both experiments. This plant is an herbaceous perennial which is native to the Mediterranean region. It is widely distributed in Italy but also in other European areas as well as in Turkey and Asia Minor. The leaves, long and strong, are still used by artisans to manufacture chairs and to produce ropes (Pasqua, Abbate and Forni, 2015, p.467).

The first technique with which the craftsperson made a basket is “sewed coiling.” The technique consists of twisting together some fibres of Ampelodesmos, thus creating a loop into which a braided cord is passed several times. It goes round and, when a cord ends, the residue is twisted with the rest of the Ampelodesmos to block it, and then the new joining thread is sewn with the needle to start again with the braid.7  For these steps, both needles worked properly, especially at points where the plot was wider. However, the sewing was not very easy for the needle with the quadrangular proximal part, while the activity was performed without any problem with the other needle. With both needles, however, it was possible to create the base of a basket (See Figure 10a, b).

The second technique used to make baskets is the “sewed-braid coiling.” This technique consists of creating the braids with seven rows of Ampelodesmos and sewing these together using the needles. In this case, the craftsperson made braids with a very wide warp. Again, both needles proved to be suitable for this technique. In fact, the wide proximal part did not interfere at all with the successful completion of the work (See Figure 11a, b, c).

Results and discussion

The shape of these specific tools is very important in order to understand how they were used. Attempts to use them as sewing needles for leather proved ineffective because of their proximal part, which is too large compared to the rest of the tool. This created many problems in the perforation of the hide. It is therefore clear that their use as sewing needles in this specific activity must be excluded.

This supports my hypothesis that these tools could be needles used in basketry, for which both experimental replicas proved sufficiently effective. Similar tools, although in metal, are still used in basket production and have morphometric characteristics similar to those of the archaeological bone needles.

It must be admitted that the shape of the replica of the needle with a quadrangular proximal part, in particular, turned out to be suitable only for a type of basket with large warps, in other words, only for coarse baskets. The slots of the archaeological needles are small and, therefore, cannot make the rope for the production of baskets like the modern ones, but this did not influence the success of the work. The size of the hole requires the use of very soft vegetables, such as the Ampelodesmos or fine cords, with which a coarse basket could possibly be made.8

The obtained result thus supports the preliminary traces analyses. In the future, more data will be obtained from the detailed analysis of the traces resulting from the experimentation presented here which will provide even more data for this study.9

Conclusion

The study of the Aradetis Orgora needles from an experimental point of view highlighted some aspects of the handicraft activities of the Bronze Age populations in the Shida Kartli region of Georgia.

From the start, I had some doubts about the suitability and use of the much larger proximal part of these tools compared to the rest of the instrument. In fact, similar tools in other chronological or geographical contexts were classified as pins for clothes, but the preliminary trace analyses carried out on them showed that they were most likely used in specific craft activities such as basketry.

The use of bone tools and needles in basketry activities is well attested both on the archaeological and ethnographic levels (Van Gijn, 2016). For this reason, an experimental study on the use of these instruments was carried out, thanks to which the possibility that they could have been used as basket needles was tested. Future paleobotanical analysis of the Aradetis Orgora site and studies on the functional traces of this experiment will allow to have more precise and precise data on the type of vegetation used for the craft.

An alternative hypothesis could have been that these needles, or some of them, may have been used in other activities, such as weaving, and perhaps used together with a loom. Actually, the two hypotheses are not mutually exclusive as, in fact, basketry is older than weaving, and techniques and tools were handed down from one activity to the other. This alternative hypothesis will be verified in future experimentation and study.

Be that as it may, our preliminary results are also important since they highlight that the site of Aradetis Orgora hosted a population with a structured economic organisation, which used specific tools-kits in different craft activities which were perpetuated within the site itself. This implies that knowledge of the raw materials and manufacturing techniques was inherent to the cultural background of these people, whose tradition of production and use of the artifacts is recognisable with a clear continuity in the population(s) that inhabited the site throughout the Bronze Age.

Acknowledgements

I would like to thank Giovanni Morra without whose help this study would not have achieved the results presented here. A further thanks goes to the operators of the Parco Regionale Riviera di Ulisse where Giovanni Morra was able to carry out the experimentation work. My acknowledgements also go to prof. Elena Rova, prof. Iulon Gagoshidze and all the staff of the Georgian-Italian Shida Kartli Archaeological Expedition of the Georgian National Museum of Tbilisi and “Ca’ Foscari” University of Venice, thanks to which I have been able to study the Aradetis Orgora objects and carry out this research. Finally, I would like to express my gratitude to prof. Cristina Lemorini, coordinator of the LTFAPA of Sapienza University of Rome, who always supported us with her precious advice.

This research was funded by a MAECI 2020 contribution for supporting the activities of the Archaeological Expedition of Ca’ Foscari University in Georgia.

• 1All the experimental protocols and the traces studies were processed and performed within the Laboratory of Technological and Functional Analyses of Prehistoric Artefacts (LTFAPA) of Sapienza University of Rome, directed by the professor Cristina Lemorini.
• 2The experimental reference trace is part of the writer’s experimental comparison collection acquired for a previous study.
• 3The choice of the bones was made according to the availability of raw material.
• 4The use of a pump drill is anachronistic in respect to the Bronze Age, but the action is the same as in other mechanic drills; see Ilan, 2016.
• 5To avoid the overlapping of the traces the author has abraded the surfaces of the objects with sandstone in order to erase the previous traces formed with the processing of the leather.
• 6The paleobotanical remains are still under study and for this reason has been chosen to use for experimentation a common plant in the Mediterranean basin but also present in Asia Minor.
• 7Detailed information on the production techniques were supplied by craftsperson Giovanni Morra during the experimentation process.
• 8All the technical knowledge about the relationship with the baskets and the use of needles were provided by artisan Giovanni Morra during the experimental activity.
• 9Trace analysis is still ongoing.