Funded by Deutsche Forschungsgemeinschaft (DFG) – project number 389351859 
http://gepris.dfg.de/gepris/projekt/389351859

Start of work: March 2018

PIs: Prof. Dr Heinrich Härke (Universität Tübingen, Germany)

        Dr Azilkhan Tazhekeev (Korkyt-Ata State University of Kyzylorda, Kazakhstan)

Project description (2017, revised March 2019):

This cooperative project explores the process of early medieval urbanization on the northern Silk Road in the context of nomad state formation and long-distance trade. The key site is the deserted town of Dzhankent which is recorded in Arab sources of the 10th to 12th centuries as the ‘capital’ of the Oguz nomad empire. Key questions include the date of the foundation of the town and its predecessor settlement; its lay-out and functional elements; the composition, structure and changing economic activities of its population; and the reasons for its demise. These questions are tackled with an interdisciplinary geoarchaeological approach by non-destructive prospection, excavation and enviromental studies in the interior of the town and in its immediate surroundings, and by analysis of finds and findings. The project is intended to give an impetus to the study of early medieval urbanization in western Kazakhstan by introducing new methodologies, and by putting the subject for the first time in the context of theories and debates from Central Asia to Western Europe.

The significance of the site derives from its historical context as well as its natural environment: the formation of the Oguz nomad state in the 9th/10th centuries and its contemporaneous urbanization, with its main towns located in the huge river delta of the Syr-Darya. One of the main stimuli for urban development here may have been the location at the intersection of the Northern Silk Road (running east-west in the Syr-Darya corridor) and the north-south trade with slaves, furs and livestock to Central Asia. The role of Khorezmian middlemen in this trade appears to be reflected in the pottery, domestic architecture and urban planning of Dzhankent. Its location gives the project an added environmental relevance in that it contributes to the wider study of the desertification of the Aral Sea region. Environmental elements of our research programme will include the dating of river delta palaeochannels close to Dzhankent, and scientific analyses of soils, animal bones and fish remains from the site.

The project addresses, thus, socio-economic as well as environmental factors in the analysis of the urbanization process east of the Aral Sea. The results are to be interpreted within the wider framework of Central Asian and European debates about urban origins and functions in early medieval state formation; these have so far been virtually unconnected debates.

Site Director of Excavations: 
Dr Irina A. Arzhantseva (Moscow)

Science collaborators of the project:

Prof. Igor Modin (Moscow – geophysical prospection)

Prof. Andrej Panin (Moscow – geomorphology)

Dr Maria A. Bronnikova (Moscow – soil science)

Dr Ashleigh Haruda (Halle, Germany – animal bones)

Dr Alicia Miller (Jena – stable isotopes)

International Advisory Committee:

Prof. Jörn Staecker† (Tübingen)

Prof. Grenville Astill (Reading, UK)

Prof. Jonathan Shepard (Oxford, UK)

Dr Luke Treadwell (Oxford, UK)

Dr Oleksii Komar (Kiev)

Dr Veronika Murasheva (Moscow) 

Prof. Zholdasbeg Kurmankulov (Almaty, Kazakhstan).

Co-opted 2018:

Dr Yurij Karev (Paris)

Dr Agusti Alemany (Barcelona)

Dr Natalia Khamaiko (Kiev)

Ph.D. students working on Dzhankent topics:

Gulmira Amirgalina (Tübingen – pottery)

Zhetesbi Sultanzhanov (Tübingen – urban planning and domestic architecture)

Sejdaly Bilalov (Almaty – fortifications)

Geophysical prospection was continued, and now covers about two thirds of he interior. Our geomorphologists and soil scientists carried out coring inside the town and in the vicinity, looking for palaeochannels of the Syr-Darya where a river port of the town may have been located. Within the town, their cores showed that in all trenches the occupation layers have a depth of about 4 metres or more – a finding that may necessitate a re-thinking of the fieldwork strategy.

New trenches were started in the citadel (no. 6), inside the west gate (no. 8) and outside the northern wall (no. 9); existing trenches in the lower town (no. 1) and at the citadel wall (no.2) were continued and extended. Trench no. 8 produced new insights into the end of the town: this did not come about by destruction, but by abandonment after which water accumulated inside the town’s enclosure and building walls collapsed. According to the latest pottery finds from the citadel, this is likely to have happened early in the 12^th century. The new radiocarbon dates just coming in (February/March 2019) seem to confirm the beginnings of the settlement in the 6^th or 7^th century. The 2018 fieldwork has produced no new evidence that would suggest a change of the 10^th century date for the fortifications (and thus, the date for the urban lay-out of the town).

In the second week of September, the project’s International Advisory Committee convened at the site for a workshop to discuss excavation techniques and the results obtained so far. On the final day, the workshop moved to Kyzylorda State University for a discussion of current interpretations, hypotheses, and prospects for further work at Dzhankent.

Excavations and related fieldwork were continued this year from the last week of July to the first week of September. The existing trenches in the citadel (Trench 6), on the citadel wall (Trench 2), and in the lower town (Trenches 1 and 8) were extended and taken deeper, providing new information on domestic architecture and the stratigraphy of the town walls. Geophysical prospection now covers all of the interior, providing a plan of the town in the 10^th – 11^th centuries and pinpointing a number of features (e.g. a circular trench in the centre of the town) which we intend to look at in the final fieldwork season.  

Having seen the sheer depth of occupation layers in the manually obtained cores last year, we have, supported by the advice of our geomorphologist (Prof. A. Panin), changed our fieldwork strategy by shifting the emphasis this year towards coring. A grid of 25 coring points was set out across the town where a mechanical drill (hired from a geomorphology firm from Moscow) dug cores of 10 cm in diameter and up to 8 m in depth – at each point down to natural. These cores are now being analysed by our soil scientists, and after C14 dating they will give us an outline of the entire occupation history of the site. Visual inspection of the cores on site already provided first essential insights, unfortunately (or perhaps fortunately, after all) destroying some cherished hypotheses and tentative identifications of features. 

Special finds in 2019:

This year, we found and excavated again a large number of modern (probably Soviet-period) burials in our trenches, more than ever before. These are a challenge for several reasons, but mostly because they are assumed to be Muslim burials, which means that they cannot be kept for later analysis: they have to be buried on the present-day village cemetery immediately after the end of the annual season. This year we were lucky to have on site a Tübingen student with relevant expertise, Sandra Reininghaus, who excavated, inspected and documented a total of 16 graves and the human bones from them. Her observations of malnourishment and stress indicators appear to lend support to our assumption that these may be burials of famine refugees of the early 1930s. 

Two other Tübingen students, Raija Heikkila and Robert Selan, paid particular attention to the processing of the numerous animal bones. This resulted in the discovery of several worked bones and a number of markings on vertebrae and knuckle bones, suggesting that they may have been used for games. Incidentally, our animal bone specialist, Dr Ashleigh Haruda, had discovered last year the bones of a cat in the stores of finds from 2014. Following this, she put together an international and interdisciplinary team for palaeopathological, genetic, stable isotope and radiocarbon analysis of the bones. Their typescript (submitted for publication in November 2019) highlights that this is the first domestic (or hybrid?) cat from a settlement context in Central Asia. With a C14 date around AD 800, Dzhanik (as we now call the cat) actually pre-dates the 10th century lay-out of the town by almost a century – and this is significant because cat-keeping was essentially an urban phenomenon, it was not a nomad habit. (see attempto article)

Some outstanding discoveries were also made in 2019, notably a large bronze vessel in a house in Trench 8, in the lower town. But the most surprising, even sensational find of 2019 was made after the end of the fieldwork season. One of our conservators, Elena Pshenichnova, discovered in one of the Dzhankent pottery vessels on her workbench a ‘nest’ of three eggs – with Arabic script on them! These are now with a specialist in the Kunstkammer Museum workshop at St Petersburg, and we eagerly anticipate the decipherment of the texts on them once her restoration work is completed. Perhaps that will help to explain the nature of this exceptional deposit against the outside of the town wall, below the citadel. 

Publications and reports about our Dzhankent project have generated a good deal of interest among foreign specialists, and members of the team have been kept busy in 2019 following invitations to conference and lecture presentations in Russia, Kazakhstan and China – and last not least, Tübingen. 

After two seasons had been cancelled because of COVID restrictions, it was a relief to return to the site this year for the third fieldwork season of the Dzhankent Project (1 August – 6 September 2022). We had not been lazy in the meantime, of course: there had been a steady stream of publications (see below), as well as contributions to online conferences, workshops and seminars (incl. Silk Roads seminar, Cambridge, UK; Nottingham University Tri-campus seminar, both in 2021). Unfortunately, face-to-face lectures and seminars in Reading, London and Uppsala in early 2022 had to be cancelled – an indirect consequence of the Russian ‘special military operation’ against the Ukraine. Conservation and laboratory work on finds (pottery, glass, animal bones, etc.) continued without much interruption although the obstacles to travel and face-to-face meetings affected some of that work.

All the greater our joy that we were able to assemble again an international team of ‘old’ specialists, young archaeologists and students in the summer of 2022. It had been our plan from the outset to devote this, probably the last, excavation season to focus on open questions. Given the number of qualified trench supervisors, we were able to continue work in the key sections: Trenches 2 (junction of citadel and town walls), 6 (citadel interior) and 8 (shahristan/lower town, just inside the west gate). These provided again important finds and essential insights into the internal structure and sequence of the site: elaboration of the key stratigraphic sequence for the site (Trench 2); details of residential occupation in the citadel (Trench 6); and burnt roof timbers and iron fittings of a large object (as yet unidentified, in Trench 8). But particular attention was devoted to less well understood features, and to issues of continuing debate within the project team.

Thus, we opened a new section (Trench 10) on the massive clay platform against the inside of the northern town wall. The existence of this feature was only realized with the help of a new Digital Terrain Model (by our geophysicists) and after deep coring in 2019 (by our geomorphologists and soil scientists). On top of this platform, we were surprised to find a sequence of substantial buildings of the 9th – 10th centuries (a preliminary date based on pottery finds), including the charred remains of what looked in situ like a burnt Schwellenbau. If further analysis and comparison confirms this finding as a sleeper beam construction, it would possibly be the first of its type of this period in Central Asia. And perhaps we have here, on this high, solid clay platform above the shahristan, the hint of a solution for the puzzling lack of monumental buildings on the citadel. In the other ‘problem’ trench (No. 11), what had looked like a circular feature on the plan of geomagnetic anomalies (now complete for the interior of the town) turned out to be far too large to have been a yurta basis (which had been our initial speculation), and it was not exactly circular either. But we found here an unexpected mass of broken, mostly complete storage vessels of impressive size. Our attempt to solve a riddle had simply opened up a new conundrum.

Elsewhere, our soil scientists collected large series of soil samples to test our conflicting speculations on the purpose of the northern annexe (animal paddock or slave pen?). These samples will be tested for parasites (at Oxford) and biomarkers (at Jena), with both approaches being able to distinguish between the presence of animals and humans. In Trench 1, in the middle of the lower town where natural had been reached earlier, the profile was sampled for a complete stratigraphic C14 sequence of the kind we already have for Trench 2 at the edge of the citadel. The previously open question of the date of the east wall and gate (which are much better preserved than all other walls of the town) got a surprise answer: first half of the 13th century! It therefore looks as if Dzhankent saw partial re-occupation and re-fortication some time after its abandonment in the 11th century, with this final phase (Dzhankent III) being contemporaneous with the construction of the Golden Horde town of Mon-tobe about 4 km away. Perhaps the deserted Oghuz town with its (probably still intact) defences was re-used as an outlying fort guarding access to Mon-tobe from the east.

Finally, it is worth mentioning that the process of ‘museification’, with reconstructions on site as well as improved information and access for members of the public, is making steady progress in a separate project run by our local Kazakh colleagues from Kyzylorda State University (which, incidentally, celebrated its 40th anniversary this year). A few weeks after the end of the 2022 fieldwork, our Dzhankent team put in a strong showing at the Vth International Congress of Archaeology of the Eurasian Steppes held at the historic town of Turkestan (or Türkistan, southern Kazakhstan) in October 2022, presenting a total of five Dzhankent-related papers on methods, results and finds of our project.