80 (129)

154 The Viking Age in Denmark

Figurc 44 Distribution of Arabie coins in a sam ple of north-west Russian silvcr-hoards

ccntury sonie Arabie coins reached western Europę from Scandinavia, found, for instance, in English silver hoards ofthe ‘Norse’ provinces.^2fJ

The Kaupang harbour apparently did not survive the late ninth-century decline in silver, and Birka certainly did not live out the drastic mid-tenth-century depression.²⁷ This shows how fragile the Scan-dinavian townships to the north of Denmark were, and also that they functioned mainly as markets for luxuries, first of all of western origin, in cxchange for Arabie and some west European silver too. At Birka a town wali was set up shortly before the site was abandoned in 950-960. The rampart is lying on top of a grave with a coin from about 925 and may reflect an inerease of raiding in the Baltic, corresponding to the tenth-century silver decline (cf. the contemporary town wali of Hedeby). The early tenth-century silver peak is reflectcd in the Birka graves, where 30 Arabie coins are of the period 890 to 930; 20 of these are from 900 to 920, while only three are latcr, from 930 to 950 (Fig. 42). The corresponding western coins are only five in number, but datę from the period 900 to 930 and are parallel to the major Arabie importation. Eight Nordic coins of the tenth century have also been found, all of which were struck at Hedeby, where minting was resumed in the early tenth ccntury after a cessation of about fifty years.

Figurę 45 Distribution of Arabie coins in a sam ple of south-west Russian silver-hoards

As in the ninth ccntury, wc notę a correlation between the periods of International tracie and coin importation from both the east and the west and the size of the silver stock, as established for Denmark. This is not to deny that most of the silver was eastern; from western Europę came morę goods.

The pattern of imported Arabie silver directs our attention to Russia. From there we have a sample of 42 hoards, with Arabie coins covering the whole Viking period from 780 to 1040 without intervals and spread geographically from the Finnish Bay to the old provinces of Kiev and Poltava and, to the east, from the area at Kazan on the middle Volga to the lower Don.²⁸ Fifteen of the finds are from before 900, 17 from the tenth century, and the rest later. The Arabie coins of the sample, from 700 onwards, totalling five thousand, display a elear two-peak distribution w^rith a deep depression between the periods of intensive import (Figs. 44-6). The first concentration, as in the Birka graves, lies around 800 - in Russia perhaps a little earlier than at Birka. The second, and strongest, peak comes after 890, ending in the tenth century, while very few coins stem from the period 870 to 890. Indeed, after 820 the number of coins is already declining in Russia. To the east, in the Kazan area, the tenth-century depression takes place as early as the second quarter of the century, while in north-west Russia, closer to eastern Scandinavia (and Birka), a morę gradual decline is seen by 950. In the south-west, however, Arabie coins still arrive in relatively large numbers until the last quarter of the tenth century.²⁹

The drastic fluctuations in the flow of Arabie silver into Russia can hardly be explained by shifting intensity of minting in the Islamie world.³⁰ They must be connected with alterations of the tradc patterns and, in the end, of the Arabie demand for, or ability to acquire, those