On the Question of Sasanian Presence in Sogdiana. Recent Results of Excavations at Paykand

Journal of Inner Asian Art and Archaeology 7/2012 Produced under the aegis of the Institute of the Ancient World, New York University by FG Founded by Lilla Russell-Smith and the Circle of Inner Asian Art, School of Oriental and African Studies, University of London editors Judith A. Lerner Sören Stark Annette L. Juliano editorial advisory board Michael Alram (Vienna), Jan Bemmann (Bonn), Henri-Paul Francfort (Paris), Angela F. Howard (New Brunswick, NJ), Lilla Russell-Smith (Berlin), Eleanor G. Sims (London), Nicholas Sims-Williams (London), Roderick Whitfield (London) We, the editors, are happy to announce two important changes relating to the editing and format of this journal. First, it is our great pleasure to announce that with this volume Annette L. Juliano has joined us as Co-Editor. Annette is an internationally respected art historian specializing in the art of China and Eastern Central Asia, with a deep focus on Buddhist art. We cordially welcome Annette and look forward to working with her. At the same time we sincerely thank Lilla Russell-Smith, the “founding-mother” of this journal, for more than ten years of valuable collaboration. It is not least due to her extraordinary scholarly standing and in-depth expertise that this journal has quickly gained worldwide scholarly recognition. The second change pertains to the general format of future volumes. In coordination with Brepols, we have decided to switch from a journal format with yearly issues to a series of edited volumes with the new title, Inner and Central Asian Art and Archaeology [volume number]: New Research. This will give us more flexibility with regard to the publication date of individual issues. Further, the new format suits us better as editors in our attempt to conceive and publish individual volumes with a stronger thematic, geographical, and/or chronological focus. We hope to achieve this in particular through the publication of relevant conference proceedings – something that we have done already in the past (volumes 2 and 3), and in this volume, which contains important contributions from the 2010 roundtable, “Shifting Frontiers: Current Issues in the History of Early Islamic Central Asia,” held on held on December 17-18, 2010 at Leiden University, guest-edited by Petra Sijpesteijn (Leiden University) and Étienne de la Vaissière (EHESS). It will now become easier for us to accommodate such contributions. This change, however, does not mean that our new series of edited volumes will give up the former JIAAA’s broad geographical and chronological approach, covering the vast region flanking the ancient Silk Roads from the Iranian world to western China, and from the Russian steppes to north-western India, with an emphasis on cultural interchange among these regions. Our commitment to this tradition is reflected in the title that we have chosen for this and the future volumes: Inner and Central Asian Art and Archaeology, to be complemented with volume-specific subtitles. We hope that this change will help us as to maintain the high academic standard of the JIAAA and at the same time serve even better what has become in recent years an ever-growing and highly diverse scholarly community. Thus, we welcome the submission of relevant scholarly articles. Prior to submitting manuscripts, authors should send a proposed title and abstract and obtain from one of the editors the revised “Notes to Contributors” explaining editorial procedures employed by the ICAAA series; these may be requested via email from Judith A. Lerner (judith.lerner@nyu.edu), Annette L. Juliano (alj328@nyu.edu), or Sören Stark (soeren.stark@nyu.edu). Manuscripts must be submitted in electronic format, including illustrations, which may be low-resolution or included in a Word document. When the article is accepted for publication, please send your hi-resolution images to Dr Judith Lerner (judith.lerner@nyu.edu via www.wetransfer.com). Contributors are responsible for obtaining from the publishers, authors, and/or owners written permission to reprint all illustrations, tables, and long text quotations taken without substantive change from sources protected by copyright, including materials not previously published. © 2016, Brepols Publishers n.v., Turnhout, Belgium. All rights reserved. No part of this publication may be reproduced stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of the publisher. D/2016/0095/182 ISBN 978-2-503-54348-2 Table of Contents Luca M. OLIVIERI, “The Last Phases at Barikot: Urban Cults and Sacred Architecture. Data from the Spring 2013 Excavation Campaign in Swat” p. 7 Shaymardankul A. RAKHMANOV†, “Wall Paintings from Tavka, Uzbekistan” p. 31 Aleksandr NAYMARK, “The Coinage of Nakhshab during the First–Fourth Centuries ce. Towards a New Systematization of Sogdian Coinages and the Political History of Sogd during Antiquity” p. 55 Andrey V. OMEL’CHENKO, “On the Question of Sasanian Presence in Sogdiana. Recent Results of Excavations at Paykand” p. 79 Sirodzh D. MIRZAAKHMEDOV, “The Ribāṭ-Caravanserais from the Eastern Suburbs of Paykand. Archaeological and Historical Aspects” p. 109 Alison BETTS, Vadim N. YAGODIN†, Frantz GRENET, Fiona KIDD, Michele MINARDI, Melodie BONNAT, and Stanislav KHASHIMOV, “The Akchakhan-kala Wall Paintings: New Perspectives on Kingship and Religion in Ancient Chorasmia” p. 125 Ken PARRY, “Reflections on a Silk Fragment from Toyuq: Christian or Manichaean?” p. 167 Tatjana BAYEROVA and Gerald KOZICZ, “The Last Wall Standing. A Survey of the Architectural Remains and the Material Components of the Dilapidated Buddhist Temple of Chigtan in Lower Ladakh” p. 193 Papers from the Roundtable “Shifting Frontiers: Current Issues in the History of Early Islamic Central Asia,” held on December 17–18, 2010 at Leiden University Petra SIJPESTEIJN (Leiden University), Étienne DE LA VAISSIÈRE (EHESS), Introduction p. 207 Minoru INABA, “Between Zābulistān and Gūzgān: A Study on the Early Islamic History of Afghanistan” p. 209 Simone MANTELLINI, Serena DI CUGNO, Rita DIMARTINO, and Amreddin E. BERDIMURADOV, “Change and Continuity in the Samarkand Oasis: Evidence for the Islamic Conquest from the Citadel of Kafir Kala” p. 227 Boris. I. MARSHAK† and Valentina I. RASPOPOVA, “Panjikent and the Arab Conquest” p. 255 Asan I. TORGOEV, “New Data on the Islamization of South-Western Semirech’e” p. 277 Addresses of Authors p. 289 Color Plates p. 291 5 On the Question of Sasanian Presence in Sogdiana. Recent Results of Excavations at Paykand* a ndrey v. o Mel’ chenko In memoriam Grigorii L. Semenov (1950–2007) INTRODUCTION Paykand was an old city, situated at the southwestern border of the Bukhara oasis in Uzbekistan; it is attributed in the historical and archaeological literature to Western Sogdiana (Fig. 1). The site is located on the lower reaches of the Zerafshan river, at a distance of two caravan stages from the capital Bukhara and another two stages from Āmul, the old and most important crossing point of the Amu-darya river (Fig. 2). In this area meet several important trade routes: one coming from Eastern Turkestan, Chāch (now Tashkent) and Central Sogdiana (Samarkand), and leading towards Margiana (Marw) and further on to the Near East, and one, coming from Eastern Europe and Khwārezm and leading south towards Tokhāristān and India. It is not by chance that during the early medieval period, i.e. in the seventh and eighth centuries ce, merchants from Paykand took an active part in the mediatory trade between China and Iran. Paykand is mentioned by many medieval authors. They report that Paykand was older than Bukhara and that it was well fortified; this last feature earned the city the name, “Brazen city” (shārestān-e rūīn).1 Interestingly, Ferdowsī mentions a “Brazen castle” (rūīn-dezh or dez-e rūīn) in his Shāh-nāma in connection with the third campaign of the Iranian prince Esfandīār against the Turanians.2 Already Josef Markwart (Marquart) identified this castle with Paykand (Markwart 1934, p. 159, 164).3 Also according to the Shāh-nāma, Ferēdūn – the legendary Iranian king and ruler of Iran and Tūrān – constructed one of the oldest fire temples in Paykand (then called Kunduz).4 Although legendary, this evidence attests to ancient contacts in this part of the Bukhara oasis with Iran (Eastern Khorāsān) and the importance of Paykand in the spread of Zoroastrianism in Sogdiana. Journal of Inner Asian Art and Archaeology 7 (2012) · 79-107 A BRIEF HISTORY OF ARCHAEOLOGICAL INVESTIGATIONS OF PAYKAND The first archaeological investigations at the site were carried out by Lev A Zimin, secretary of the Turkestan Circle of Enthusiasts of Archaeology in 1913 and 1914. In 1939 and 1940 excavations were conducted by the Zerafshan Expedition, organized by the State Hermitage and the Institute for the History of Material Culture (both in what was then Leningrad) and with the participation of the Uzbek Committee for the Protection of Monuments of Antiquity and Arts (UzKomStarIs). Systematic investigations, however, started only in 1981 with the work of the Bukharan Archaeological Expedition, jointly organized by the State Hermitage Museum and the Institute of Archeology of Uzbekistan, which in 1983–85 was joined by the expedition of the Moscow Museum of Oriental Art. With respect to its assigned task – the investigation of ancient and medieval urbanism in Central Asia in all its manifestations – all constituent parts of the former city were, over the following 35 years, subjected to systematic investigation: the citadel (ark), an entire living quarter in both the so-called Shahrestan-1 and the Shahrestan-2, a bāzār at the southern gate of the city, the street-grid, its fortifications, two suburban areas,5 and other objects (Fig. 3). To date, c. 11 % of the total of 20 hectares of the settlement has been studied. The results of this work have been published in a number of monographs, numerous articles, and, since 1999, in the form of comprehensive annual reports of the expedition.6 THE LATE ANTIQUE CONTEXTS FROM THE CITADEL In connection with a special research program, initiated in 2003, to study the most ancient stages of the establishment of the city, considerable excavation has been focused on Paykand’s citadel (ark). These investigations made it clear that the citadel acquired its current configuration in the third-fourth centuries ce (Figs 4 and 5). The citadel occupies an area of c. one hectare in the northeastern corner of the site. For the first time, material dating to the third-fourth centuries ce and earlier, were obtained in the stratigraphic trench that was dug in the area of the so-called “palace” (dating to the early Middle Ages) in the northeast of the citadel (Mukhamedzhanov et al. 1983, pp. 148–56). Due to the presence of massive early medieval and pre-Mongol deposits on the citadel, sizeable areas yielding Late Antique contexts could be exposed only during investigations at the citadel’s fortifications, and at the temple in the north of the citadel. Temple The temple of Paykand has been discussed both in a book and in several articles by Semenov, who identified four main construction phases in its history (Semenov 1996b, pp. 37–44). Excavation revealed that during the third-fourth centuries ce (Stage 3 according to our recent periodization7), the center of the complex contained two sanctuaries (7 × 5.3 and 6.9 × 5.5 m), each surrounded by corridors. In the center of each sanctuary stood a podium, on which presumably portable fire altars were placed; the floors of the sanctuaries were covered with alabaster plaster. Semenov identified numerous early analogies for this type construction: the Oxus temple, Surkh Kotal, Dilberjin, temples in Susa and Kuh-e Khwaja (Semenov 1995, pp. 172–74). This identification of the building as a temple, however, met with criticism from Boris I. Marshak, the renowned expert on Sogdian history, archeology, and art.8 Subsequent work in 2003– 07, however, revealed additional evidence in favor of Semenov’s identification by revealing a 17 × 12 m courtyard in front of the sanctuaries during the phase 3 (=old period 2)9 and later (Omel’chenko 2008, pp. 85– 90), based on a c. four-meter high platform, built of pakhsa and mud brick; additional remains of an earlier platform could be traced inside this platform (Semenov, Mirzaakhmedov et al. 2005, fig. 6). The entrance to the courtyard (whose width varied across the periods from between 1.7 and 0.9 m) was situated in the eastern wall, the direction of sunrise. Moreover, the design of a tablelike construction, 1.15 × 0.46 m with a niche (filled with ashes), was recognized; this construction probably served in some purification ritual performed upon entering the temple. New excavations also discovered a lateral two-flight 80 stairway,10 leading from the courtyard toward up to the eastern sanctuary (Fig. 6). It was built of sun-dried mud bricks as well as of fired bricks of an unusual trapezoid shape (60 × 24.32 × 6.54 × 17.25 × 5 cm), which also covered the wall. It was plastered with clay and alabaster.11 South of the platform and at a slightly lower level we found the so-called “outer courtyard,” c. 3.5 m wide. This courtyard was subsequently used to dispose of rubbish from the temple – mainly consisting of аsh, pieces of charcoal, and fragments of alabaster plaster from the ṣuffa benches as well as the floors of the sanctuaries. The ceramic inventory from this spot is limited to goblets, cups, jars, hand-made cauldrons, pans, and incense burners. In addition, a great number of animal bones were found, including those belonging to such wild animals as boar and roe deer.12 In the filling were found three large daggers and five knives, together with numerous smaller fragments of knives or daggers. Earlier, Semenov had found inside the sanctuaries twelve knives and one dagger, as well as more than 200 small, standardized cups. All of this is distinct evidence of cultic ceremonies – connected with religious festivals – that were carried out in the temple.13 Additionally, ceramics of the first century ce and imitations of Antiochus coins with a horse head on the reverse (belonging to E. V. Zeimal’’s third period of emissions were found in the lower layers of the outer courtyard (Zeimal’ 1983, p. 85, tab. 9, 060–074). At Stage 4 – the latest construction phase of the temple – the entrance to the upper courtyard was transferred to the south, and the lateral staircase was turned into a ramp (pandus). At Stage 5, the space of inner courtyard, filled with waste and remains of construction, was leveled by a platform of mud brick,14 on which a four-column ayvān with a wide central staircase leading to the sanctuaries was constructed (Semenov, Mirzaakhmedov et al. 2007, 8, fig. 6; Semenov, Mirzaakhmedov et al. 2000, fig. 3, 8.). To the south of the platform a lower area (16. 5 × 7. 8 m) was formed. Thus, even during the latest phase of the existence of the temple is retains its tripartite structure: cella and two courtyards, each one located at different levels (Fig. 7). Geographically and chronologically, the closest comparanda to the Paykand temple is that of Yerkurgan, the old capital of the region of Nakhshab in the lower Kashka-darya area.15 Here, also, an ayvān of similar dimensions (18.5×7 m) was located in front of the sanctuaries, facing south and rising above a vast courtyard; during period 1, this courtyard was constructed also in the form of a terrace. Additional commonalities shared by both temples are the raising of the sanctuaries and the ayvān on a platform, the monumentality of the walls, staircases, and a special area, in form of a small courtyard, for the deposition of the remains of the sacred fire, together with numerous animal bones, ceramics, and votive items (Suleimanov 2000, pp. 90–93, 98, 102–06, fig. 41). An interesting parallel for the non-central location of the staircase leading from the inner courtyard up to the platform of the sanctuary is provided by temple I of Panjikent. Here, the platform with the ayvān, the cella and the surrounding corridors could be accessed from the inner courtyard through a centrally located, gently raising ramp.16 However, during period 2 of temple 1 at Panjikent, at the east corners of the platform (facing the inner courtyard), small stairs provided additional access to the platform. It seems that the southern one of these staircases allowed communication between the ātashgāh (south of the platform) and the sanctuary (cella), perhaps in order to transport ember from the ātashgāh to the cella.17 Two lateral stairs, giving access to corridors surrounding ayvān and cella, accompanied the central ramp also at the temple of Dzhar-tepa-2.18 In Iran, an early example using lateral stairs in addition to a central one is the so-called āyadana of the Achaemenid period in Susa.19 There are other details of the plan of the Paykand temple of Stages 3–4 that find parallels in sanctuaries within the wider Iranian world. These are, as examples, the typical vertical organization of space with the help of terraces and staircases/ramps,20 the pairing of columns in the ayvān, the obstruction of sight of the rooms with altars (deliberately breaking axes of direct access), and the use of alabaster plaster and fired bricks in the interior. The specifics of the temple of Paykand lie in the existence of two sanctuaries connected by side corridors. There were also two sanctuaries in Surkh Kotal, but they were separated from each other by a courtyard. It has been suggested that they were used during different parts of the ritual. Two rooms with an ayvān connected by a surrounding corridor were in the so-called postAchaemenid “Frataraka temple” near Persepolis.21 Two pairs of altars of post-Achaemenid date can be found in Naqsh-e Rustam (Litvinskii and Pichikian 2000, pp. 272– 73, 328–29, 237–39, fig. 57, 58, 3). But the perhaps the closest similarity is the main core of temple of the Sasanian royal fire, Ādur Gushnasp, at Takht-e Suleiman. Here we find the chahārṭāq (A) and the ātashgāh (B) connected and surrounded by corridors.22 Thus, based on its plan of sanctuaries, having corridors on three sides with entrances to the cella, the temple of Paykand belongs to an architectural type, widespread since Parthian times (Litvinskii and Pichikian, 2000, p. 205; Suleimanov 2000, p. 250). Military Architecture Also our work on the fortifications of the Paykand citadel revealed contexts dating to Late Antiquity. Semenov’s excavations exposed the entire northern defensive wall and its segments in the northeastern and northwestern corners (Fig. 5). These fortifications evolved during several construction periods. The main construction phase is represented by Stage 1 at the northern flank and partially at the western flank, along with Stage 2 at the eastern flank.23 The northwestern sector was strengthened by five square towers with arrow slits, each c. 18–20 cm wide. The eastern sector of the northern defensive wall was without towers, but the northern corridor was, at the upper floor, divided into compartments. This stage was dated between the end of third and the first half of the fourth century (Semenov 1996a, p. 104, 140). In the southeast of the citadel, synchronous layers were identified in the late 1990s to early 2000s when excavations revealed segments of a long corridor, and approximately the middle of the southern defense line, the so-called “room with high ṣuffa benches” (Semenov, Mirzaakhmedov et al. 2000, 23, fig. 28, 29, 32; Semenov, Mirzaakhmedov et al. 2004, 8, fig. 100). In 2004, work in the southwestern sector of the citadel began.24 The earliest complex revealed here was a chain of rooms, adjacent to the inner flank of southern fortification wall, which were all accessed from a joint corridor (the so-called “corridor-comb layout” (in Russian “коридорогребенчатая планировка”; Fig. 8).25 All rooms were of nearly the same size (c. 4.5 × 3.5 m), all almost entirely surrounded by ṣuffa benches, 0.8–1m wide; such ṣuffas were also found in the corridor. Both on the floors and on the ṣuffa benches of the rooms were traces of impermanent fireplaces. Two rooms (5 and 17) were not designed for habitation, but appear to have been used for storage as numerous fragments of large storage vessels – so-called khums – were found, along with pits for their placement in the floor (Fig. 9, 2). These rooms are, along the southern and western suite of rooms respectively, the fifth ones, counted from the corner room (room 9). This, together with other details, speaks in favor of the existence of a pre-conceived, overall plan for the entire complex prior to construction of this complex of rooms. These rooms were covered by flat ceilings, resting on wooden beams with a diameter of c. 20–25 cm, and set 81 c. 1.05–1.15 m apart from each other; the height of their ceilings was originally 3.65–3.75 m (Fig. 9, 1; 10, 1). Partition walls between the rooms were very thin (namely half a brick, i.e., 30–35 cm). Poles and reeds lay across the beams, strapped by bundles of grass (which left imprints on the floors when the ceilings collapsed), and plastered with clay. Two bases of sandstone were found in room 8, which may indicate the presence of a hatch in the ceiling to provide access to the second floor via a wooden ladder. Arrow slits, 18–20 cm wide and 40 cm high, were cut out of the rooms through the fortification wall of citadel. The top row of arrow slits was located at a height of 1.3 m from the original floor, allowing for a man to stand while shooting, while the bottom row was 0.6 m, to shoot by kneeling. Taking into account the angle of inclination of the arrow slits, this attests to an average height of over 1.70m for the archers. The well-preserved interior wall, separating the suite of rooms from the corridor, continued up to one meter above the beam holes for the first floor ceilings; thus, there was a second floor. But, because no mud-brick rubble was found in the rooms, the second floor was probably an open defense area. It may have been covered by light awnings under which, at some places, vessels stood because fragments of khums were found on the collapsed remains of the ceiling in three of these rooms. The walls continued upward further above the ceiling of corridor, which allows us to conclude the presence of a chemin de ronde – a raised protective walkway – above the corridor. Thus, defense of the citadel in the southwestern corner was effected by at least three means and levels: the arrow slits, the parapet over the walls, and the parapets over the corridor. Several more details of the rooms are unusual for ordinary dwelling spaces. The partitions between the rooms are only partially covered with plaster, only up to a maximum height of 90 cm; this is convenient only for someone sitting on the ṣuffa. In seven rooms, the walls near to the entrances were covered with a series of incised vertical lines, varying from 12 to 24, crossed by a horizontal line. Possibly, they reflect a method of counting, such as the number of days of duty). The width of the southern aisle of the corridor, into which the southern rooms opened, was 1.87 m, while the one of the western aisle was 2.36 m. The entrances from the corridor were 1.0–1.2 m wide and 2.2 meters high, had two steps, and were aligned to the left partitions of the rooms. The top of the entrance openings were mostly arches, semicircular in shape, with mud bricks, placed horizontally and on their short side with an arch brick on 82 top. In some cases we could identify remains of horizontal wooden grooves from the original doorframes. Our stratigraphic investigations showed that the rooms along the fortress wall contained three to four layers of floor plaster. Under the southern face of the fortress wall and the room behind it sat a platform, constructed of 11 rows of mud bricks, beneath which was virgin sand. The situation was different at the western face of the fortress wall where a sondage in room 15 revealed a platform made of 22 layers of mud brick, for a height of at least 4.6 m. This difference in height may be explained by the slope of the natural hillock on which Paykand’s citadel was originally built. The outer edge of the platform was excavated in the southwestern corner of the citadel. Here the fortress wall reached a thickness of up to 2.5 m at its foot (Fig. 10). With a width of 5.5 m for the platform, this portion of the citadel was well-protected from siege machines. Some of the arrow slits in the bottom row had their upper parts in the fortress wall, while their lower parts sat directly upon the platform. The arrow slits show a varying angle, perhaps depending on the relief in front of the platform. It is noteworthy that some of the lower parts of the arrow slits that sit in the platform were actually blocked by the partitions walls between the rooms. This gives us some glimpse into the sequence of construction sequence: first, the platform was erected, with the lower part of the arrow slits in the bottom row indented, after which the fortress wall was built. Subsequently, two long walls were constructed parallel to the fortress wall. These resulting corridor walls featured, at standardized distances, arched openings. Afterwards, the space between the fortress wall and the outer wall of the corridor was divided into a suite of rooms. Construction of the complex proceeded from east to west, i.e., from the entrance to the citadel; and apparently this took some time. On the uppermost bricks of the platform and the construction floor, covering them (before the partitions between the rooms were erected), we found not only ceramic fragments but also sand and wind-deposited clay which must have accumulated at this spot when the rooms were already standing, but still were without roofs. As revealed by the excavations of this complex, a number of details coincide with those previously observed in other parts of the citadel. For example, we also found a platform under the southeast end of the southern axis of the corridor. Its width is 3 m, the same as that of the archer corridor at the northern fortification wall. The thickness of the latter is 2.6 m, which, in turn, is almost the same as that of the southwestern fortification (2.5 m). The distance between the fortification wall and the wall of the corridor – i.e., the space which was later divided into a suite of rooms – is 3.7 m, which is exactly the same as the width of the corridor in the eastern sector in the southwest. Further, we find in all parts the same types of masonry, as well as the same format of mud bricks and the same kinds of findings. In my opinion, all this points to a substantial rebuilding of the fortresses in the area of Paykand’s later citadel, following a general master plan. The structures were considerably expanded to the south, while the new northern walls sat on the older ones. With that, the final extension of the later citadel was reached. This seems to have facilitated two purposes: first, it considerably strengthened the defense potential of the site, and, secondly, made it possible to station a sizeable garrison. The previously mentioned “corridor-comb plan” had been typical for barrack structures at least since late Antiquity (see Khmel’nitskii 2000, p. 67). Thus, it is beyond doubt, that with their standardized interiors and settings, the suites of rooms in the southwest corner of Paykand’s citadel fortifications should be interpreted as barracks. Their location close to the only entrance into the citadel – which makes it the weakest part of the fortress – is only logical. Judging by the comparatively low height of its archaeological deposits (0.5–1 m), the barracks in the southwest of the citadel seems to have functioned only for a short period. The rooms were systematically dismantled and abandoned: the ceiling beams were removed, the arrow slits closed or simply blocked by fragments of brick or large shards from khums; passages leading from the corridor into the citadel were closed with masonry. Gradually, the corridor and the rooms at the fortification wall were filled with blown sand. After some time (the sand layer had accumulated to c. 0.5 m high) two towers were built at the center of the southern façade of the fortress, so that they flanked the entrance to the citadel (Figs 7, 21). The western tower was built on the collapsed walls of a former barrack room (Fig. 22). A bent corridor was connected with the tower, leading from the entrance deep into the citadel. Ṣuffa benches and traces of habitation inside the tower suggest that the tower room (room 1/6) housed the guardians of the gate. To this second construction period in the southwest of the citadel also belongs the reuse of room 17 of the former barracks as a storage place for khums. At about the same time reuse the northwestern corner tower of the citadel also begins (Semenov 1996a, p. 104, 140). Chronology and Character of the Material Assemblage These significant changes in the military architecture of the Paykand citadel correlate with a massive transformation of the material complex, as clearly reflected our findings; these, in turn, allow us to date this construction phase with considerable accuracy, the main evidence provided by coins found on the floors and in the layers directly above them. Numismatic remains from settlements in the Bukhara oasis indicate that until the first century bce small-scale monetary transactions in the wider Bukhara area were based on small silver imitations of Antiochus silver coins with a horse head on the reserve, while from the first to the third centuries ce the function of small change passé to the Hyrcodes series.26 All the coins serving small trading transactions found in the Paykand barracks, however, were copper – a phenomenon completely alien to the previous and other contemporary coinages in Sogdiana.27 From the lower floors – the third or the fourth – of the complex came copper imitations of late Kushan coins, issued by Vasudeva and Kanishka II (III) (Fig. 11, 9–16), which circulated in Tokhāristān from the second half of the third century and during the whole of the fourth, well after Kushan rule was ended by the Sasanians’ victories. On the two upper floors of the barracks these coins were found in association with Kushano-Sasanian coins (Fig. 11, 2–8), which circulated at the beginning of the fourth century in territories ruled by largely autonomous viceroys of the Sasanians in the East (Gorin 2009, 19). Based on these finds, the upper floor of the barrack complex should be dated to the first third of the fourth century. Also important to note is that above the upper floor, in the filling of room 7, was a silver imitation of an Euthydemus tetradrachm with the legend MR’Y (Livshits and Lukonin 1964, p. 169; Fig. 11: 1). It belongs to the first variant within Group 2 of L. N. Kazamanova’s typology (1961, pp. 121–22, 127; pl. 2, 11), or to Michael Alram’s Group 5 (1986, 275; pl. 39: 1228, 1229). Such coins were minted in Bukhara in the second half of the second or the first half of the third century (Naymark 2008, pp. 62–68) and judging from their weight (the Paykand example is 8.48 g), they served for large trading operations (Omel’chenko 2011, pp. 47–54). Ceramics reveal three stages (Fig. 12). The most numerous and diverse group comes from the upper floor; this can be explained by the large scale of excavations and the massive thickness of the layer above this floor (and 83 probably also by the broader chronological framework of Stage 3). Although vessels of all three phases demonstrate a general uniformity in terms of technology and typology, there are some differences from phase to phase that, in my opinion, indicate an increased southern influence on the ceramic repertoire of Paykand during Late Antiquity. In section, the ceramic fabric is bright red in color. As a result of firing under high temperatures, almost all storage vessels have a characteristic pale surface;28 only a small percentage of grey ware is represented. Almost half of the entire corpus consists of fragments of hemispherical cups. Their rim is sometimes slightly thickened and they often feature an earlier and thus uncharacteristic ring foot (Fig. 12: I, 6–10. II, 3–10; III, 10–14). Cups, as well as some ewer forms, display a direct dependency on the Kushano-Sasanian ceramic complex of Tokhāristān, and, in particular, with phase II/2 of Vladimir A. Zav’alov’s typology (2008, pp. 79– 81). Interestingly, among the ceramic finds in Paykand is a particular type of cup with hemispherical body of relatively large diameter and with thickened rim, on which a small notch has been made (Figs. 12: III, 15). But the clearest marker for the ceramic complex presented here are the so-called taghāra bowls with twisted handles, interior ledges and ribs inside, and incised wavy and dotted patterns (Figs. 12: III, 19-20). They were found in the upper habition layers of the barracks. Similar taghāra bowls have been also found in Marw and Khwārezm (Vainberg et al. 1981, fig. 44, 17), although they are most typical for Kushano-Sasanian complexes in Northern Bactria/Tokhārestān of the third and first half of the fourth century (Zav’yalov 2008, p. 194: 84, 6). We must also mention some “northern” influence in the handmade pottery at Paykand during this period, especially noteworthy in examples of the so-called “barbecue stands” (Russian: шашлычницы) that feature an image of ram’s (bull’s) head (Fig. 13, 21); also attested are lids, pots, and censers on legs, decorated with impressions fingers and small relief bands, and sometimes an incised “the tree of life.” Obviously, this pottery complex is associated with the so-called Kyzyl-kyr group in the Bukhara oasis. It has been suggested that this material belongs to the western wing of the tribes moving into the oases of Sogdiana from the Syr-Darya region (Kaunchi-1, Otrar-Karatau and Dzhety-asar cultures).29 Among the individual findings from the barrack rooms with clear analogies in the Kushano-Sasanian complex of Northern Bactria-Tokhārestān, from the end of the third and the fourth century, we should also mention bone hairpins with figural finials (Fig. 14, 17-18), marble84 like limestone spindle whorls decorated with carved concentric patterns (Fig. 14, 10-11), and clay horse figurines.30 Stamped on the base of one of the ceramic spindles is a swastika (Fig. 14, 16); similar items were found archaeologically in the territory of the Kushan state. Characteristic for the Late Antique period are large beads made of blue glass paste with white eye beads. Unexpected are finds of fragmentary glass vessels, which rarely occur during so early a period at sites north of the Amu-darya. The body of one of these vessels – apparently, from a goblet or a cup – was decorated with soldered-on glass strips (Fig. 14, 15). This kind of decoration is well known from Late Antique Roman glassware, examples of which spread far to the east (Kropotkin 1970, p. 30). Perhaps connected Western (Eastern Roman) glyptic art is the impression on a molded lid, executed with a seal with a serrated rim. It features the bust of a curly-haired man wearing a necklace (Fig. 13, 9). On a burnt clay bulla from the fill of room 14 comes a male bust with high “Parthian” hairstyle, supported by a diadem (Fig. 14, 1). On the lower floor of the archers’ corridor, to the north of the citadel, synchronous with the structures of the barracks in the southwest, we found, along with ceramics, a small group of iron weapons, which includes two daggers (one with a ring pommel), a knife with protruded backside, a two-bladed spearhead, and three-bladed arrowheads (Fig. 15); a similar arrowhead was found in the barracks. This type of arrowhead was widespread in Western Central Asia in the third-fifth centuries ce.31 As mentioned above, the end of this construction phase is marked by a hiatus, followed by a phase of disintegration. The next phase of construction on the fortifications areas should be dated to the second half of the fourth or the beginning of the fifth (?) century ce. This is suggested by the ceramic complex (Fig. 23) – which is not very different from the one found in the former barracks – and the find of a Kušānšāh coin (presumably Bahrām II, around 350 ce) from the disintegration layer above the floor of the corridor in the north of the citadel (Semenov 1996a, p. 138). Of interest is the find in this context of a terracotta figurine depicting the so-called “Bukharan goddess” (Fig. 23, 1).32 Wall Paintings In 2015, a ground-penetrating radar survey confirmed the existence of an old entrance in the citadel’s southern front, between room 1/6 of the barracks and the so-called “room with high ṣuffa benches” (Fig. 16). The layout and interior design of this room allowed us to suggest in 2003 that it belonged to the gateway complex, leading inside the fortress – although its south front was closed by a wall, built in the following construction period. The small group of ceramics coming from the top floor of this room is obviously synchronous with the material from the barracks (Fig. 17; Semenov, Mirzaakhmedov et al. 2004, pp. 112–13). Renewed excavation in 2015 revealed under the third floor of the “room with high ṣuffa benches” a pathway paved with of fired bricks (Fig. 18). This masonry continued under the later wall, blocking passage to the south, towards the entrance to the citadel. At the eastern edge of the pathway remains of collapsed adobe walls were discovered which preserved portions of wall painting (Fig. 19). The largest fragment (40 × 45 cm)33 features three figures, their shoulders partially overlapping each other. Since only their torsos are preserved, it is impossible to determine if they are walking or standing. The figure in front wears a kandys-like upper garment with open semicircular collar and fringe, decorated with pearls, and holds some object in his right hand, reminiscent of a torch or a censer (only the shaft of the object is preserved). The second character also wears a robe. His arm bends at the elbow and his palm faces forward. Only the right hand of the third person is preserved and it holds what may be a mace. The painting features white, yellow, pink, and red colors. On the upper part remain spots in a blueblack color, apparently, the remains of hair or beards. Curiously, the hands are painted in bright red. The same color was used to depict the hands of an adorant at the mural fragment, discovered in 1998 nearby, namely on the wall of the corridor (?), leading to the palace. But stratigraphically, this fragment is situated much higher, and the accompanying material assemblage suggests a date in the sixth or – possibly at the end of the fifth century ce (Semenov and Adylov 2006, pp. 36–37, fig. 2). Procession scenes and rows of figures standing next to each other often occur in Sogdian painting, the bestknown example being the painting from the so-called “palace of Ikhshēds,” Sector 23/Room 1, at Afrasiab or Old-Samarkand (Al’baum 1975, see, for instance, figs 4, 5, 8). But in the case of the new fragment from Paykand the position of the figures is unusually close – overlapping each other so that only their right shoulders are visible. This compositional device is vividly reminiscent of the rock reliefs of the Sasanian Shāpūr I at Naqsh-e Rajab (Fig. 20a) and Bīshāpūr (Fig. 20b). In this context, it should be mentioned that parallels with Sasanian art have already been noted in a wall painting at the site of Uchkulakh in the northwest of the Bukhara oasis, excavated by the Uzbek-Italian archaeological expedition (Silvi Antonini and Mirzaachmedov 2009; Lo Muzio 2010). In a room dating to the end of the fifth or the beginning of the sixth century, were preserved traces of a mural showing a jousting scene (?), reminiscent of several Sasanian reliefs (Lo Muzio 2014; Lo Muzio 2015, pp. 109–18). In my opinion, the new Paykand fragment, which is now one of the earliest known examples of Sogdian painting suggests that the influence of Sasanian art on that of Sogdiana dates back to the end of the third or the beginning of the fourth centuries. HISTORICAL CONTEXT AND CONCLUSION As I hope to have shown, we now have substantial evidence to suggest that the fortress at the site of the later citadel of Paykand was significantly expanded and enhanced at some point during the second half of the third century ce. Further, the evidence from the barracks in the southwestern corner of the fortress indicates that this new design was followed a well-conceived master plan. This, of course, raises the question: for whom and why were these large-scale constructions undertaken at the southwestern periphery of the Bukhara oasis? To begin to answer this question, it is, first of all, necessary to recall that these constructions included the fire temple. Interestingly, the plan of the temple’s central core is quite different from the sanctuaries of other Sogdian temples; however, the plan has close parallels to fire temples in Iran – specifically, the central part of that of Takht-e Soleymān, which possesses a čahārṭāq and a ātašgāh. Secondly, we find many elements in the material culture of the so-called Kushano-Sasanian complex widespread in those territories of the former Kushan kingdom that were seized by the Sasanians (Fig. 24). Thirdly, there appears to be a clear influence of Sasanian or Kushano-Sasanian visual art, recognizable in the above-mentioned recently discovered wall paintings. And finally – and most important – cardinal changes occur in the local coinage. Now, in the Bukhara area, for the first time in Sogdiana, copper coins now circulate; previously, only silver coins have been recognized there. This change is particularly key to understanding the building program at Paykand. Almost three decades ago Aleksandr Naymark argued that finds of Kushan imitations and Kushano-Sasanian coins in Paykand and in the Kum-Sovtan oasis testify to the Sasanian occupation of this area (Naymark 85 1987). He saw the confirmation of this conclusion in the introduction of a bi-metallic coinage with Sasanian inspired iconography that took place in Bukhara in the early fifth century (Naymark 1995). Naymark also tried to mobilize further archaeological and epigraphic evidence suggesting that Bukhara was a part of the Sasanian realm for a protracted period of time (Naymark 1990; Naymark 2001, pp. 69–72). In my opinion, this suggestion can now be confirmed, made more specific, and expanded. The Bukhara region, indeed, came for a certain time under the political control of the Sasanians, with the fortress of Paykand becoming an important stronghold for them. This might well have happened already under Šāpūr I. In his inscription on the Kaʾba-ye Zardošt, he speaks about his realm extending over “Kušān-šahr tā frāz ō Paškabūr ud tā ō Kāš, Sugd ud Čāčestān” – “Kušān-šahr bis vor Pešāwar(?) und bis Kāšghar(?), Sogdien und Taškent.”34 With the discovery of the rock relief of Rag-e Bibi in northern Afghanistan, most likely depicting the same king hunting an Indian rhinoceros,35 it has become obvious that the Kaʾba-ye Zardošt inscription accurately reflects the political situation, atleast at the eastern borders of the Sasanian empire. If we turn to the north, there are essentially two interpretations of the inscription for Kāš (MPers and Parthian) / Κας (Greek): Kašghār (in presentday Xinjiang), or Keš (the upper Kashka-darya region in present-day Uzbekistan; Lur’e 2004, 10, n. 11). In either case, the possessions of Šāpūr I reached into Sogdiana. Indeed, the border between the former Kušān realm and the various Sogdian principalities controlled by Kangju 康居 ran along the Hissar range, separating the Surkhandarya valley (Northern Bactria) from the Kashka-darya (Southern Sogdiana; Masson 1968, pp. 14–25). After the Sasanians had seized Kušān-šahr constellation seems to have remained. But then how could the troops of Šāpūr I have reached the borders of Čāč, an area that lay far to the north, beyond the lands of Sogdiana? For an explanation one needs to look at Sogdiana not from the southeast, but from the southwest. If the region of Bukhara was indeed subordinated to Šāpūr I,36 this border indeed exists; it runs along the western limits of the Nuratau range in the northeast of the Bukhara oasis. This mountain range represents the natural watershed between the Zerafshan and Middle Syr-darya, the latter representing the historical region of Čāč. Thus the Kaʾbaye Zardošt inscription seems to contain fully adequate evidence. Of course, the political situation in western Central Asia during the expansion of Sasanian Persia was highly complex. In particular, as shown by the Paykand 86 materials, already in the third century ce, the migration of semi-nomadic groups began from the Syr-darya region into the oases of Sogdiana. Perhaps it was against these groups (or against those who forced them to migrate) that the rulers of Samarkand, Keš, Nakhshab and Bukhara tried to resist by having the “leader of the people of Čāč” (the Tashkent region) construct a fortress at the frontier with the nomads, as indicated by the Kul-tobe inscription (Podushkin 2005, pp. 133–39; Sims-Williams and Grenet 2006, 106–07). Already in the 340s, the Sasanian Šāpūr II waged intense wars in the northeastern parts of his empire with the Chionites (who were coming, perhaps, from the Syrdarya region). Later these Chionites appear as allies of Iran in its wars with the Roman Empire Ammianus Marcellinus XVI 9.4, XVII 5.1, XVIII 6.22, XIX 1.7–11, 2.1–6). It is likely that the Sasanians lost their control over the Bukhara oasis during these tumultuous years, and, finally, also had to relinquish their stronghold at Paykand. Subsequently, however, Iran continued to exert cultural and probably political influence on the Bukhara region. In any case, the reverse design of the copper coins of Mawāk, the first independent local ruler of Bukhara, as well as the famous Bukhārkhudā drachms imitating the coins of Bahrām V, clearly testifies to this. Of course, our suggestions concerning the history of the relationship between the Sasanian realm and the Bukhara region are preliminary and need further substantiation. Fortunately, the site of Paykand continues to furnish us with new evidence. Thus, only during our field season in 2014 excavations in a new area at the citadel revealed a coin of Šāpūr I (used already in antiquity as a pendant) – so far the only one found in an archaeological context beyond the Oxus river (Fig. 25). ABBREVIATIONS IMKU Istoriia matrial’noi kul’turi Uzbekistana, Tashkent-Samarkand MBAE Materiali Bukharskoi Expeditsii. St. Petersburg ONU Obshestvennie Nauki Uzbekistana, Tashkent RA Rossiiskaia Arkheologiia, Moscow SAI Svod Arkheologicheskikh Istochnikov, Moscow TGE Trudi Gosudarstvennogo Ermitazha, St. Petersburg TIuTAKE Trudi Iuzhno-Turkmenistanskoi Arkheologicheskoi Expeditsii, Ashkhabad TKhAEE Trudi Khorezmskoi arkheologo-etnograficheskoi ekspeditsii, Moscow VDI Vestnik drevnei istorii, Moscow BIBLIOGRAPHY Al’baum 1975 Lazar I. Al’baum, Zhivopis’ Afrasiaba [Paintings of Afrasiab] (Tashkent: Fan, 1975). Ambartsumian 2013 Artur A. Ambartsumian. “‘Mednaia krepost’ v pekhleviiskoi traditsii [The ‘Cupper fortress’ in the Pahlavi tradition],” in Commentationes Iranicae. Vladimiro f. Aaron Livschits nonagenario donum natalicium, ed. Sergei R. Tokhtas’ev and Pavel’ B. Lurje (Sankt-Peterburg: Nestor-Istotiia, 2013), 229–48 Askarov et al. 1991 Akhmadali A. Askarov, Edvard V. Rtveladze, and Galina V. Shishkina, Kultura i iskusstvo drevnego Uzbekistana [Culture and Art of Ancient Uzbekistan]. Exhibition catalogue, 1. (Moscow: Vneshtorgizdat, 1991). Baratova 2001 Larisa S. Baratova. “Mednye monety s izobrazheniem luchnika i ikh mesto v monetnom chekane Iuzhnogo Sogda [Copper coins with an image of an archer and their place in the coinage of Southern Soghd],” IMKU 32 (2001): 143–46. Baratova 2004 Larisa S. Baratova. “K periodizatsii denezhnogo obrashcheniia v Iuzhnom Sogde (domusul’manskii period) [Towards the periodization of the monetary circulation in Southern Soghd (pre-Islamic period],” in Transoxiana. Istoriia i kul’tura [Transoxania. History and Culture], ed. A. Saidov et al. (Tashkent: Institut Otkrytoe Obshchestvo, 2004), 181–88. Berdimuradov and Samibaev 1992 Amriddin E. Berdimuradov and Masud K. Samibaev. “Rezultati raskopok khrama na Dzhartepa II [The Results of Excavations on the Temple of Dzhartepa II],” IMKU 26 (1992): 77–92. Biruni – Sal’e 1957 Biruni, Abu Reikhan. Pamiatniki minuvshikh pokolenii. Izbrannye proizvedeniia [The Remaining Signs of Past Centuries. Selected works]. Translation and notes by Mikhail A. Sal’e (Tashkent: Fan, 1957). Boyce and Grenet 1991 Mary Boyce and Frantz Grenet. A History of Zoroastrianism. Volume Three: Zoroastrianism under Macedonian and Roman Rule (Leiden et al.: Brill, 1991). Callieri 2007 Pierfrancesco Callieri. L’archéologie du Fārs à l’époque hellénistique. Quatre leçons au Collège de France, 8, 15, 22 et 29 mars 2007 (Paris: Éditions de Boccard, 2007). Ferdowsi – Banu-Lakhuti 1965 Abu ‘l-Qasim Ferdowsi. Shahnameh (in six volumes). 3. Translated into Russian by Tsetsiliia B. Banu-Lakhuti, notes by A. Azer and Tsetsiliia B. Banu-Lakhuti (Moscow: Publishing House of the Academy of Sciences of the USSR, 1965). Ferdowsi – Banu-Lakhuti 1969 Abu ‘l-Qasim Ferdowsi. Shahnameh (in six volumes) 4. Translated into Russian by Tsetsiliia B. Banu-Lakhuti, notes Vladimir G. Lukonin (Moscow: Publishing House of the Academy of Sciences of the USSR, 1969). Gorbunova 2000 Natal’ya G. Gorbunova. “O vooruzhenii sredneaziatskikh skotovodov (II (III?) v. do n. e.–V v. n. e.) [On the Weapons of Central Asian Nomads (2nd–3rd) c. bc–the 5th c. ad],” Rossiiskaia Arkheologiia 2 (2000): 40–51. Gorin 2009 Alexei N. Gorin. Denezhnoe obrashchenie v Severnoi Bactrii v seredine II v. do n. e.–III v. n. e. [Coins Circulation in Northern Bactria in the middle of the 2nd c. bc–the 3rd c. ad] (Dissertation abstracts, Tashkent: 2009). Huyse 1999 Philip Huyse. Die dreisprachige Inschrift Šābuhrs I. an der Kaʾba-i Zardušst (ŠKZ) (London: School of Oriental and African Studies, 1999). Isamiddinov and Suleimanov 1984 Mukhamadzhon Kh. Isamiddinov and Rustam Kh. Suleimanov. Erkurgan (stratigrafiia i periodizatsiia) [Erkurgan (stratigraphy and periodization)] (Tashkent: Fan, 1984). 87 Kazamanova 1961 Liudmila N. Kazamanova. “Podrazhaniia tetradrakhmam Evtidema v sobranii GIM [Imitations of tetradrachms of Euthydemus in the collection of the State Historical Museum],” Vestnik Drevnei Istorii 3 (1961): 120–27. Khmel’nitskii 2000 Sergei Khmel’nitskii. Mezhdu kushanami i arabami. Arhitektura Srednei Azii 5–8 vv. [Between Kushans and Arabs. Architecture of Central Asia of the 5th–8th centuries ad] (Berlin, Riga: Gamajun. 2000). Kropotkin Vladislav V. Kropotkin. Rimskie importnie izdeliia v Vostochnoi Evrope (II v. do n. e.–V v. n. e.) [Roman imported wares in Eastern Europe (2nd c. bc–5th c. ad)] (SAI D1–27, Moscow: Nauka, 1970). Livshits and Lukonin 1964 Vladimir A. Livshits and Vladimir G. Lukonin. “Srednepersidskie i sogdiiskie nadpisi na serebianykh sosudakh [Middle Persian and Sogdian Inscriptions on Silver Vessels],” Vestnik Drevnei Istorii 3 (1964): 155–76. Litvinskii 2001 Boris A. Litvinskii. Khram Oksa v Baktrii (Iuzhnii Tadzhikistan). 2: Baktriiskoe vooruzhenie v drevnevostochnom i grecheskom kontekste [The Temple of the Oxus in Bactria (Southern Tajikistan). 2: Bactrian weapons in the Ancient Eastern and Greek context] (Moscow: Vostochnaia Literatura, 2001). ed. Deborah Klimburg-Salter and Linda Lojda (Turnhout: Brepols, 2014), 225–36. Lukonin 1974 Vladimir G. Lukonin. “Zavoevaniia Sasanidov na Vostoke i problema kushanskoi absoliutnoi khronologii,” in Tsentralnaia Aziia v kushanskuiu epokhu. Trudi Mezhdunarodnoi konferentsii po istorii, arkheologii I kul’ture Tsentralnoi Azii v kushanskuiu epokhu, 1 [Sasanian Conquests in the East and the Problem of Kushan absolute chronology, in: Central Asia in the Kushan Epoch. Proceedings of the International Conference on the History, Archaeology and Culture of Central Asia in the Kushan Period, 1], eds Bobodzhon G. Gafurov, Grigorii M. Bongard-Levin, and Edvin A. Grantovskii (Moscow: Nauka, 1974), 302–07. Lur’e 2004 Pavel B. Lur’e. Istoriko-lingvisticheskii analiz sogdiiskoi toponimii [Historical and Linguistic Analysis of the Sogdian toponymy] (Dissertation abstracts, St. Petersburg, 2004). Lur’e 2015 Pavel B. Lur’e. “Toponimicheskie dannie o bukharskom dialekte sogdiiskogo iazika,” in Bukharskii oazis i ego sosedi v drevnosti i srednevekov’e [Toponymic Data on Bukhara Dialect of Sogdian Language, in: Bukhara Oasis and Its Neighbors in the Ancient and Medieval Times], eds Andrei V. Omel’chenko and Dzhamal K. Mirzaakhmedov (“TGE” 75, St. Petersburg: State Hermitage Publishing House, 2015), 127–42. Litvinskii and Pichikyan 2000 Boris A. Litvinskii and Igor R. Pichikyan. Ellinisticheskii khram Oksa v Baktrii (Iuzhnii Tadzhikistan). 1: Raskopki, arkhitektura, religioznaia zhizn [The Hellenistic Temple of the Oxus in Bactria (Southern Tajikistan). 1: Excavations, architecture and religious life] (Moscow: Vostochnaia Literatura, 2000). Lurje 2006 Pavel B. Lurje. “‘Shapur’s Will’ in Bukhara,” in Ēran ud Anērān: Studies presented to Boris Ilich Marshak on the Occasion of His 70th Birthday, ed. Compareti Matteo, Raffetta Paola and Scarcia Gianroberto (Venice: Cafoscarina, 2006), 407–18. Lo Muzio 2010 Ciro Lo Muzio. “An Archaeological Outline of the Bukhara Oasis,” Journal of Inner Asian Art and Archaeology 4 (2010), 43–68. Markwart 1938 Josef Markwart. Wehrot und Arang. Untersuchungen zur mythischen und geschichtlichen Landeskunde von Ostiran (Leiden: Brill, 1938). Lo Muzio 2014 Ciro Lo Muzio. “New Evidence on Sogdian Painting from Uch Kulakh (Bukhara Oasis – Uzbekistan),” in South Asian Archaeology and Art I: Changing Forms and Cultural Identity: Religious and Secular Iconographies, Masson 1968 Mikhail E. Masson. “K voprosu o severnikh granitsakh gosudarstva Velikikh Kushan [On the problem of northern borders of the Great Kushans],” ONU 8 (1968): 14–25. 88 Mukhamedzhanov et al. 1983 Abdulakhad R. Mukhamedzhanov, Rustam Kh. Suleimanov, B. Urakov. Kultura Drevnebukharskogo oazisa III–VI vv. n. e. [The Culture of the Ancient Bukhara oasis of during the 3rd to the 6th centuries ad] (Tashkent: Fan, 1983). Mukhamedzhanov et al. 1988 Abdulakhad R. Mukhamedzhanov, Shukhrat T. Adylov, Dzhamal K. Mirzaakhmedov and Grigorii L. Semenov. Gorodishche Paikend. K probleme izucheniia srednevekovogo goroda Srednei Azii [The city-site of Paikend. On the problem of the study of medieval cities in Central Asia] (Tashkent: Fan, 1988). Naymark 1987 Aleksandr I. Naimark. “Drevnie monety iz KumSovtana [Ancient Coins from Kum-Sovtan]” in Vtoraia Vsesoiuznaia Numizmaticheskaia Konferentsiia (Moscow, 1987), 50–51. Naymark 1990 Aleksandr I. Naimark. “K numizmaticheskoi istorii Nakhsheba [On the numismatic history of Nakhshab],” in Arkheologiia Srednei Azii. Tezisy dokladov (Tashkent: Tashkent State University, 1990), 66–68. Naymark 1995 Aleksandr I. Naymark. “O nachale chekanki mednoi moneti v Bukharskom Sogde [On the beginning of minting of copper coins in Bukhara Sogd],” NTsA 1 (1995): 29–50. Naymark 2001 Aleksandr Naymark. Sogdiana, Its Christians and Byzantium. A Study of Artistic and Cultural Connections in Late Antiquity and Early Middle Ages (PhD thesis Indiana University) (Bloomington: 2001). Naymark 2008 Aleksandr I. Naimark. “O prichinakh poiavleniia podrazhanii ellinisticheskim monetam v Sogdiane [On the reasons for the appearance of imitations of Hellenistic coins in Sogdiana],” Vestnik drevnei istorii 1/264 (2008): 55–70. Naymark 2011 Aleksandr I. Naymark. “An Elusive Stage in the Samarqand Archer Coinage,” Journal of the Oriental Numismatic Society 208 (2011): 39–46. Obel’chenko 1974 Oleg V. Obel’chenko. “Kurgannie mogilniki epokhi kushan v Bukharskom oazise,” in Tsentralnaia Aziia v kushanskuiu epokhu. Trudi Mezhdunarodnoi konferentsii po istorii, arkheologii I kul’ture Tsentralnoi Azii v kushanskuiu epokhu. Tom 1 [Kurgan barrows of the Kushan epoch in the Bukhara Oasis, in Central Asia in the Kushan Epoch. Proceedings of the International Conference on the History, Archaeology and Culture of Central Asia in the Kushan Period. Vol. 1], eds Bobodzhon G. Gafurov, Grigorii M. Bongard-Levin, and Edvin A. Grantovskii (Moscow: Nauka, 1974), 202–09. Omel’chenko 2011 Andrei V. Omel’chenko. “Podrazhaniia tetradrakhmam Evtidema s gorodishcha [Imitation of tetradrachms of Euthydemus from the site of Paikend],” NTsA 10 (2011): 47–54. Omel’chenko, Mirzaakhmedov et al. 2013 Andrei V. Omel’chenko, Dzhamal K. Mirzaakhmedov, Alexei N. Gorin, Alexei V. Kulish, Dmitrii V. Sadofeev, Nazbergen Zh. Saparov, Normamat D. Sobirov, Ruslan M. Toiirov, Asan I. Torgoev and Dilmurad O. Kholov. Otchet o raskopkakh v Peikende v 2011–2012 gg. [Report on the excavations in Paikend in 2011–2012] (“MBAE” 12, St. Petersburg: Publishing House of the State Hermitage, 2013). Podushkin 2005 Alexander N. Podushkin. “Novie pamiatniki pis’mennoi kul’turi Iuzhnogo Kazakhstana [New Records of the Written Culture from South Kazakhstan],” Shygys 2 (2005): 133–39. Rutkovskaiia 1962 Lukeriia M. Rutkovskaiia. Antichnaia keramika Drevnego Merva [Antique ceramics of Ancient Merv] (“IuTAKE” 11, Ashkhabad, 1962), 93–147. Semenov 1995 Grigorii L. Semenov. “Sviatilishche v Paikende,” in Ermitaznye chteniia pamiati V. G. Lukonina [The Sanctuary in Paikend, in The Hermitage Readings in Memoriam of V. G. Lukonin], ed. Evgenii V. Zeimal’ (St. Petersburg: Publishing House of the State Hermitage, 1995), 171–78. Shliumberzhe 1985 Daniel’ Shliumberzhe [Schlumberger]. Ellenizirovannii Vostok [The Hellenized East] (Moscow: Iskusstvo, 1985). 89 Semenov 1996a Grigorii L. Semenov. Sogdiiskaiia fortifikatsiia V–VIII vekov [Sogdian Military Archtecture of the 5th to 8th centuries ad] (St. Petersburg: Publishing House of the State Hermitage, 1996). Semenov 1996b Grigori L. Semenov. Studien zur sogdischen Kultur an der Seidenstrasse (Wiesbaden: Harrassowitz, 1996). Semenov and Adylov 2006 Grigorii L. Semenov and Shukhrat T. Adylov. “Arsenal na tsitadeli Paikenda [The arsenal on the citadel of Paikend],” in Ancient and Medieval Culture of the Bukhara Oasis, eds. Chiara Silvi Antonini and Dzhamal K. Mirzaakhmedov (Samarkand and Rome: The Institute of Archaeology of the Academy of Sciences of the Republic of Uzbekistan, 2006), 36–43. Semenov, Mirzaakhmedov et al. 2000 Grigorii L. Semenov, Dzhamal K. Mirzaakhmedov (with participation of Shukhrat T. Adylov, Igor K. Malkiel’, Alexander V. Bekhter). Raskopki v Paikende v 1999 g. [Excavations in Paikend in 1999] (“MBAE” 1, St. Petersburg: Publishing House of the State Hermitage, 2000). Semenov, Mirzaakhmedov et al. 2004 Grigorii L. Semenov, Dzhamal K. Mirzaakhmedov, Bakhtier M. Abdullaev, Igor K. Malkiel’, Normamat D. Sobirov and Asan I. Torgoev. Raskopki v Paikende v 2003 g. [Excavations in Paikend in 2003] (“MBAE” 5, St. Petersburg: Publishing House of the State Hermitage, 2004). Semenov, Mirzaakhmedov et al. 2005. Grigorii L. Semenov, Dzhamal K. Mirzaakhmedov, Bakhtier M. Abdullaev, Andrei V. Omel’chenko, Nazbergen Zh. Saparov and Normamat D. Sobirov. Raskopki v Paikende v 2004 g. [Excavations in Paikend in 2004] (“MBAE” 6, St. Petersburg: Publishing House of the State Hermitage, 2005). Semenov, Mirzaakhmedov et al. 2007. Grigorii L. Semenov, Dzhamal K. Mirzaakhmedov, Bakhtier M. Abdullaev, Andrei V. Omel’chenko, Nazbergen Zh. Saparov, Normamat D. Sobirov and Asan I. Torgoev. Otchet o raskopkakh v Peikende v 2006 g. [The Report of the Excavations in Paikend in 2006] (“MBAE” 8, St. Petersburg: Publishing House of the State Hermitage, 2007). 90 Shkoda 2009 Valentin G. Shkoda. Pendzhikentskie khrami i problemi religii Sogda: V–VIII vv. [The temples of Pendzhikent and problems of religion of Sogd: 5th–8th centuries ad] (St. Petersburg: Publishing House of the State Hermitage, 2009). Silvi Antonini and Mirzaachmedov 2009 Chiara Silvi Antonini and Džamal K. Mirzaachmedov (eds). Gli Scavi di Uch Kulakh (oasi di Bukhara). Rapporto preliminare, 1997–2007 (Pisa/Roma: Fabrizio Serra Editore, 2009). Stronach 1985 David Stronach. “On the Evolution of the Early Iranian Fire Temple,” in Papers in honour of Professor Mary Boyce. Volume II, ed. Harold W. Bailey, A. D. H. Bivar, Jacques Duchesne-Guillemin and J. R. Hinnells (“Acta Iranica” 25, Leiden: Peters, 1985), 605–27. Suleimanov 2000 Rustam Kh. Suleimanov. Drevnii Nakhshab. Problemi tsivilizacii Uzbekistana: VII v. do n. e.–VII v. n. e. [Ancient Nakhshab. Problems of the Civilization of Uzbekistan of the 7th c. bc–7th c. ad] (Tashkent, Samarkand: Fan, 2000). Tārīkh-i Bukhārā, ed. Riḍawī ²1984 Tārīkh-i Bukhārā, ed. M. Riḍawī (Tehran: ²1984) Vainberg et al. 1981 Bella I. Vainberg, Mikhail Lapirov-Skoblo, Elena E. Nerazik and Sof’ia A. Trudnovskaia. Gorodishche Toprak-kala [City-site of Toprak-kala] (“TKhAEE” 12, Moscow: Nauka, 1981). Zav’ialov 2008 Vladimir A. Zav’ialov. Kushanshahr pri Sasanidakh. Po materialam raskopok gorodishcha Zartepa [Kushanshahr under the Sasanians. On the results of excavations at the site of Zartepa] (St. Petersburg: Faculty of Philology and Arts, State University, St. Petersburg, 2008). Zeimal’ 1972 Evgenii V. Zeimal’. “Talibarzinskii klad monet s izobrazheniem luchnika [The Talibarzu Hoard of Coins with the Image of an Archer],” Soobshcheniia Gosudarstvennogo Ermitazha 34(1972): 70–75 Zeimal’ 1978 Evgenii V. Zeimal’. “Politicheskaia istoriia Transoksiany po numizmaticheskim dannym [The political history of Transoxiana according to numismatic data],” in Kul’tura Vostoka. Drevnost’ i rannee srednevekov’e, ed. Vladimir G. Lukonin (Leningrad: Avrora, 1978), 192– 214. Zeimal’ 1983 Evgenii V. Zeimal’. Drevnie moneti Tadzhikistana [Ancient Coins of Tajikistan] (Dushanbe: Donish, 1983). Notes * Translated from the Russian by Sören Stark, Shujing Wang, and Qi Xiaoyan. 1. Tārīkh-i Bukhārā, ed. Riḍawī ²1984, p. 26, 30, 61. 2. Ferdowsi – Banu-Lakhuti 1969, p. 153, 158, 160–61, 182–84, 428. 3. See, however, also Ambartsumian 2013, pp. 229–48, esp. 241. 4. Ferdowsi – Banu-Lakhuti 1965, pp. 358–59. See the discussion of this passage in Semenov 1996b, pp. 35–36. 5. On the ribāṭs in Paykand see the article of Sirodzh K. Mirzaakhmedov in this volume. 6. The last one: Omel’chenko, Mirzaakhmedov, et al., 2013. 7. Due to new data and observations, Semenov’s original periodization scheme has been somewhat altered. These changes are reflected in the annual (preliminary) reports of the expedition. Semenov’s periods 2, 3, and 4 are now considered as Stages 4 and 5, while Semenov’s first period is now considered as Stage 3 at this excavation spot. Traces of an earlier construction (Stages 1 and 2) were encountered under the sanctuaries and in stratigraphic trenches under the floors of the corridors. Our recent works have not substantially changed the plan of the sanctuary during the Stages 2 and 3. 8. See also Shkoda 2009, p. 16. 9. Judging by the presence of ceramics dating to the third-fourth centuries ce. 10. Perhaps the first flight of stairs (3.8 × 2.5–2.75 m) had the function of a sufa-bench like, for example, the one in the temple of Erkurgan (Xenippa-Nakhshab in Southern Sogd, modern Qarshi region). But there the ṣuffa-stairs connected the courtyard with the ayvān (Suleimanov 2000, p. 98). 11. The use of fired bricks and alabaster plaster in pre-Islamic Central Asia was mostly restricted to public buildings. 12. This offers additional evidence of environmental differences between the present (desert steppe) and the past (tugai forests with numerous water sources). Of course, the discovery of animal bones mixed with the remains of the sacred fire can hardly be reconciled with what we know about the ritual regulations prescribed by canonical Zoroastrianism. However, similar assemblages have been found at other sites associated with fire cult (e.g., Dahan-e Ghulāmān and Takht-e Soleymān in Iran, Yerkurgan in Sogdiana). Scholars have attributed this phenomenon to the influence of ancient local religious beliefs (Semenov 1996b, p. 49; Litvinskii, Pichikian 2000, pp. 270– 71; Suleimanov 2000, pp. 103–06). 13. In this context it should be remembered that, according to al-Bīrūnī, the Zoroastrian priests (mages) of Bukhārā used to gather during the festival of n-k-ḥ-āghām, celebrated in the month of ps’kyc (the fourth month of the Sogdian calendar) in the city of Paykand (Biruni – Sal’e 1957, p. 254). 14. Semenov, Mirzaakhmedov et al., 2007, 8, fig. 6. 15. The investigated remains of the temple of Yer-kurgan date from the beginning of the third to the seventh century ce. Five main construction periods have been identified (Suleimanov 2000, p. 88, 101–02). 16. Note that the staircase in phase 5 of the temple of Paykand might have been turned into ramp (The Report of the Excavations in Paykand in 1997, 1 [typewritten]). 17. Shkoda 2009, p. 64, fig. 11. An iron brazier, which apparently was used for this purpose, was found in a sanctuary of the Paykand temple (Semenov, 1995, p. 172, 176). 18. Berdimuradov and Samibaev 1992, p. 84, fig. 1. As the one at Paykand, the temple at Dzhar-tepa-2 was oriented to the south, but not to the east as were the temples at Panjakent. Fire temples in Iran had not strict orientation (Litvinskii, Pichikian 2000, p. 194). 19. Stronach 1985, p. 624, fig. 3, 6. The function of this building as a temple, however, is not certain. 20. Ghirshman 1976, frontispiece, fig. 16, 41, 42; Schlumberger 1985, pp. 63–64, fig. 48–50; Suleimanov 2000, p. 98, 102. 21. Again, the function of this building as a temple is not certain; see Boyce and Grenet 1991, pp. 117–18. Recently, Pierfrancesco Callieri favored again an interpretation as temple (Callieri 2007, p. 65). 22. Stronach 1985, p. 624, fig. 3, 9. The Paykand temple of period 5 had an entry in the western sanctuary (ātashgāh?); see Semenov 1996b, p. 12. 23. An earlier wall segment was revealed at the eastern flank. 91 24. Excavations here were begun by Bakhtiyor Abdullaev, continued by Asan Torgoev, and have been conducted since 2009 by the author. 40; Mukhamedzhanov et al., 1983, pp. 113–14; Isamiddinov and Suleimanov 1984, p. 152; Pugachenkova 1989, pp. 129–30. 30. Zav’ialov 2008, p. 106, 107, 111–14. 25. One room (7 × 2–3.5 m) was opened on the other side of the corridor; there were many doors that were closed by brickwork, leading originally inside the citadel. 26. Zeimal’ 1978, pp. 198–209; Askarov, Rtveladze, Shishkina, et al. 1991, catalog nos 210, 212, 230–33. 27. This change in the Bukhara coinage and its historical reasons were already observed by Aleksandr Naymark (Naymark 1995; Naymark 2001, pp. 68–70). Small silver coins continued to circulate until the sixth century in Samarkand (Zeimal’ 1972; Zeimal’ 1978, pp. 208–10; Zeimal’ 1983, pp. 268–76; Naymark 2011), while the South Sogdian principality of Nakhshab minted in the third and fourth centuries both silver and copper denominations (Naymark 1990; Baratova 2001; Baratova 2004). 28. Although some forms of tableware have similar characteristics, in many case we have to speak here of white engobe (slip). This phenomenon is, in particular, known from the ceramic complex of the second and third centuries ce from Marw – cf. Rutkovskaia 1962, p. 68. 29. Levina 1971, p. 3, 185–93, 197, 211; 59, 28, 31–34, 35– 92 31. Obel’chenko 1974, p. 207; Gorbunova 2000, p. 48; Litvinskii 2001, p. 101 and 102. 32. Terracotta figurines of this type had been found earlier in the Bukhara oasis. As the figurine from Paykand shows no traces of reduction, typical for later “generations” of figurines (generated by using consecutive generations of figurines for producing new molds), and because it was found in stratigraphic association with KushanoSasanian coins, the beginnings of the production of this type of terracotta figurines can now be dated to the fourth century ce. 33. Restoration work was carried out by Dilmurad O. Holov, restorer at the Bukhara Museum; drawing and partial reconstruction by Larisa Iu. Kulakova, architect of the Bukhara expedition. 34. Middle Persian text and translation by Huyse 1999, pp. 23– 24; see also Lukonin 1974, p. 302. 35. Grenet et al., 2007, pp. 243–67. 36. It should be noted that the name “Šāpūr” can be found in the hydronymy of Bukhara – see Lurje 2006. Fig. 1. Map. Historical Regions of Western Central Asia (drawing by author). Fig. 2. Bukhara Oasis. Schematic map (after Mukhamedzhanov 1978). 93 Fig. 3. Site of Paykand. Plan with the excavation areas. Fig. 4. Views on the Citadel. 94 Fig. 5. Citadel. Plan of excavations. 95 Fig. 6. Citadel. Temple of the third stage, lateral stairway. Fig. 7. Temple on the citadel of Paykend: 1 – stage 3 (2): a, b – sanctuaries, c – inner courtyard, d – external courtyard; 2 – stage 5: a, b – sanctuaries, c – iwan, d – courtyard (plan of the sanctuaries after Semenov, 1996. Abb. 10–14); 3 – central part of the temple in Takht-e Soleymān: A – čahārtaq, B – ātashgāh (after Stronach, 1985, Fig. 3, 9). 96 Fig. 8. Barracks at the southwest of the citadel of Paykand. Plan and architectural section. 97 Fig. 9. Barracks. 1 – dwelling rooms; 2 – storage room. 98 Fig. 10. Barracks. 1, 2 – southwestern corner, rooms and corridor; 3 – remains of the platform with loopholes. 99 Fig. 11. Coins from the barracks. 1: silver tetradrachm of Euthydemus, obverse: ruler’s head with tiara and radiant nimbus; reverse: Hercules with mace on omphalos; 2–16 – coppers. Fig. 12. Barracks. Table ware (I, II, III stages) of the second half of the third to first third of the fourth century ce. 100 Fig. 13 Barracks. Molded ware. 1–5 – fragments of incense burners, 6–9 – lids (9 – with the impression of a finger ring), 10–16 – cauldrons, 17, 18 – pans, 19 – stand of khum, 20, 21 (so-called “barbecue stands”). See Color Plate 1. Fig. 14. Single finds from the barracks. 1 – bulla, 2–9 – beads, 10–14 – spindle whorls (14 – buttons?), 15 – fragments of a goblet, 16 – bottom of a bowl (?), 17, 18 – hairpins, 19 – belt buckle, 20 – ear, 21 – cup or collar of a jug, 22 – box; 1,12,13 – clay; 2 – agate (?); 3,5–9, 15,16 – glass; 4 – coral; 10, 11 – marbelized limestone; 14, 17, 18 – bone; 19 – silver; 21,21 – bronze; 22 – iron with copper inlay. See Color Plate 2. 101 Fig. 15. Archer corridor and barracks. Iron weapons. 1–3 – arrow-heads; 4 – javelin; 5 – knife; 6 – dagger. See Color Plate 3. Fig. 16. Southern entrance to the citadel. Results of GPR-survey. See Color Plate 4. Fig. 17. Ceramics from the upper floor of the “room with high ṣuffa benches.” 102 Fig. 18. The “room with high ṣuffa benches,” level 4; pathway of burn bricks. Fig. 19. The “room with high ṣuffa benches,” fragments of mural painting of the second half of the third century ce. Photo and drawing by Dilmurad O. Kholov and Larisa Iu. Kulakova. See Color Plate 5. 103 Fig. 20. Rock reliefs of Šāpūr I (240/243–71/273): a – Naqsh-e Rajab; b – Tang-e Showgan (© I. O. Gurov). 104 Fig. 21. Remains of the tower at the entrance to the citadel, second half of the fourth century ce. Fig. 22. Constructions of the tower, destroying the archway of the entrance to the barracks. 105 Fig. 23. Citadel. Pottery assemblage and terracotta figure of the second half of the fourth century ce. Fig. 24. Analogies in the material cultures of Paikand: a. in Bukharan Sogdiana and Zar-tepe; b. in Bactria-Tokhārestān during the Kushano-Sasanian period (comparanda from Tokhāristān after Zav’ialov 2008, passim). 106 Fig. 25. Coins from Paykand. 1. Šāpūr I; 2. Mawāk; 3. Bukhār khudā drachm. See Color Plate 6. 107 COLOR PLATES ANDREY V. OMEL’CHENKO Plate 1. Barracks. Molded ware. 1–5 – fragments of incense burners, 6–9 – lids (9 – with the impression of a finger ring), 10–16 – cauldrons, 17, 18 – pans, 19 – stand of khum, 20, 21 (so-called “barbecue stands”). Plate 2. Single finds from the barracks. 1 – bulla, 2–9 – beads, 10–14 – spindle whorls (14 – buttons?), 15 – fragments of a goblet, 16 – bottom of a bowl (?), 17, 18 – hairpins, 19 – belt buckle, 20 – ear, 21 – cup or collar of a jug, 22 – box; 1,12,13 – clay; 2 – agate (?); 3,5–9, 15,16 – glass; 4 – coral; 10, 11 – marbelized limestone; 14, 17, 18 – bone; 19 – silver; 21,21 – bronze; 22 – iron with copper inlay. 300 COLOR PLATES ANDREY V. OMEL’CHENKO Plate 3. Archer corridor and barracks. Iron weapons. 1–3 – arrowheads; 4 – javelin; 5 – knife; 6 – dagger. Plate 4. Southern entrance to the citadel. Results of GPR-survey. 301 COLOR PLATES ANDREY V. OMEL’CHENKO Plate 5. The “room with high ṣuffa benches,” fragments of mural painting of the second half of the third century ce. Photo and drawing by Dilmurad O. Kholov and Larisa Iu. Kulakova. 302 COLOR PLATES ANDREY V. OMEL’CHENKO Plate 6. Coins from Paykand. 1. Šāpūr I; 2. Mawāk; 3. Bukhār khudā drachm. See Color Plate 6. 303