Le fait urbain en Asie centrale préislamique : approche diachronique, approche synchronique, III : la crise urbaine et la réurbanisation (^IIIe-VIe s.), un processus général ? (continued) (9)

This new network of forts in no way indicates a demographic decline in these periods, but rather a revival and densification, linked in part to the integration of the "Huns", a phenomenon for which we also have onomastic clues (cf. supra). The idea of demographic pressure and/or a continuing rural exodus is confirmed by the subsequent expansion of urban territories: Punjikent (late^5th-early 6th^century ?), Paykent (second half of 6th^century ?), Samarkand (6th^century ?). The average size is now 15-20 ha. Forts were also expanding, and the new ones built were larger than before. The aristocratic class seems to be growing, with its dependent chākar. In the ^6th and 7th^centuries , during the "Western Turkish" period, there was an evolution in the type of fortifications: archways were concentrated on the upper level only in Paykend, and became mostly false in Punjikent; the lower level was either filled in or transformed into a blind communication gallery; square towers were partially replaced by more widely-spaced round towers, from which we deduce a concentration of defense, horizontally towards the towers, vertically towards the upper parapet. The walls became thicker at the base, reaching 15 m at Punjikent from the time of the Arab threat, which reintroduced siege warfare to Central Asia, a practice that had scarcely been practiced since Greek times. These extremely costly reinforcements bear witness to the economic resources of Sogdian cities. In this second phase, all this suggests a professionalization of garrisons, with troops relying more on speed of maneuver than sheer mass. Semënov assumes various variations according to the period: in Punjikent in the 6th^century , Hephtalite barracks occupying an ancient rampart; in Paykend after 560, Turkish contingents delegated by the qaghan; and finally, just about everywhere, chākar, mercenary or on duty.

New urban foundations from the^{5th century} onwards

There are no contemporary descriptions. The Chinese were interested only in the lengths of the ramparts, classified into broad categories, and in Buddhist monasteries; the Arabs only in the ramparts (Tabari barely alludes to the peri-urban belts of Zoroastrian mausoleums). The Syriac version of the Roman d'Alexandre conveys an echo of the richly painted décor of the temple of the goddess Nana in Samarkand, but it is not to Sogdian painting that we should look for panoramic or even partial views of Central Asian cities. As in previous periods, archaeology remains the main source. For lack of time this year, these foundations (mainly Punjikent, now 50% excavated) will be studied in later lectures [21]. To give an initial overview, we present an extract from a 3D animation of Pendjikent produced by a Japanese team in collaboration with the excavators.

It's worth noting the sharp contrast that has now been established between urban housing (all social classes) and rural housing (known mainly from the excavation of the village of Gardon-i Hisor, in the mountains above Punjikent). The only house in the village that resembles a town house is that of the lord. This gives us three categories: the castral residence and the rich urban dwelling that transposes it; the urban craftsman's house, sometimes decorated with paintings; and the peasant's house dependent on a castle, reminiscent of the compact two- or three-cell dwellings previously found on military sites. It should also be pointed out that, from this period onwards, we encounter concordant clues to the institutional structuring of the urban community: the nāf, "people", in fact the meeting of the heads of the main families, who had the capacity to appoint and depose the king and, in Punjikent, probably met in the temples; the khwēsht "mayor" (or in Tokharestān theandagbed "head of the village"), which heralds the ra'is of the Islamic era.

[21] Some references: G. L. Semënov, Sogdijskaja fortifikatsija V-VIII vekov, op. cit; B. Marshak, "Les fouilles de Pendjikent", CRAI, vol. 134, no. 1, 1990, pp. 286-313; V. I. Raspopova, Zhilishcha Pendzhikenta, Leningrad, 1990; F. Grenet, "Crise et sortie de crise en Bactriane-Sogdiane aux ^{IVe-Vesiècles} ", in La Persia e l'Asia centrale da Alessandro al X secolo, Rome, Accademia Nazionale dei Lincei, 1996, pp. 367-390 (although I would be less positive today about a general influence of the Sassanid urban planning model).