Islamic Ivory Chess Pieces, Draughtsmen and Dice

by Anna Contadini

There is no comprehensive study of Islamic gaming pieces, and information on them is scattered in general books or specialist articles. Research on their identification, their relationship to the various games, their history, and the development of types and shapes would contribute both to the study of games and gaming pieces in Islam, and also, given that many are in ivory, to the history of Islamic ivories in general. In this context the collection of ivory gaming pieces in the Ashmolean Museum (1) is of particular importance: not only are the seventeen pieces of very fine craftsmanship (fig.I), but they are also different in style, type and date. They fall into two main groups, depending on the games for which they were used, chess or table games, and will be examined accordingly.

Chess (2)
Chess pieces during the Islamic era fall into two broad families as far as shape is concerned. In one, the pieces are more or less naturalistic representations of figures (3) while in the other they have abstract forms. When, where and why chess pieces started to become abstract is still a matter for debate. (4) It is probable that both types were already in use soon before the Islamic era, but unfortunately our knowledge of this period is very limited.(5) Setting aside the Venafro and S. Sebastiano's catacomb pieces, which are probably of tenth or eleventh-century Italian manufacture, (6) one can cite as possibly relevant only the chess set excavated at Afrasiyab and datable, according to the archaeological context, to the seventh century AD.(7) Our knowledge of pre-Islamic times thus relies essentially on literary references.(8)

In the Islamic period we have no object certainly identifiable as a chess piece before the ninth century. Possibly the earliest are those excavated by the Metropolitan Museum of Art's Iranian Expedition at Nishapur, which are now in the Museum.(9) They are in ivory and of abstract design, with virtually no decoration apart from carved vertical lines (fig.4). There is one interesting exception: the horse (Knight), even if stylized, is still recognizable as a horse. Another Knight, in stone, very similar in its stylized form, is in Kuwait, Dar in

(p.111)

(Figure 1 The Islamic ivory chess pieces, draughtsmen and dice in the Ashmolean Museum.)

al-Athar al-Islamiyya, published here for the first time (fig 5).(10) Those in other sets of abstract pieces are different, however, having a conical shape with one symbolic protuberance at the front (fig.6). But this does not necessarily mean that the shape of the Nishapur piece is archaic, even though there is a temptation to assume that the abstract Nishapur designs derive from earlier figurative styles. In the present state of our knowledge it is not possible to determine the relative chronology of these two types.

All twelve chess pieces in the Ashmolean collection are abstract, and only four probably belong to the same set. Therefore, the first problem is that of identifying the pieces. Unfortunately, in board diagrams in early Islamic treatises on chess the pieces are usually not represented, being identified only by their names, which in Arabic are: shah (King), firman (Queen), fil (bishop), faros (Knight), rukhkh (Rook) and baidaq (Pawn). (11)

It is in a Western manuscript of 1283, the treatise on chess of Alfonso X the Wise, (12) that the pieces are first found clearly drawn (fig.7), (13) with board positions enabling the correct identification to be made. The treatise illustrates in detail every phase of the game, and also the making of chessmen, which are apparently turned on a bow-lathe (fig.8). (14) Other Western manuscripts portray pieces with a (p.112) strong resemblance to those shown here. The Manesse manuscript, a collection of Middle High German love lyrics of the first quarter of the fourteenth century, contains a painting of a chess match between Margrave Otto IV of Brandenburg and a lady, with clearly drawn pieces.(15) A manuscript in Kassel, dated 1334, has a miniature of a couple playing chess.(16) Those in all three manuscripts may be assigned to style set B defined below (see p.8), except for the Bishop and Knight, which belong to set A (fig.9b).

In the Islamic world, later illustrated treatises on chess contain accurate representations of boards and positions, but the pieces are not represented, again only the names being given. They are sometimes shown on the board in miniatures from the Shahnama, illustrating the passage on how the game passed from India to Persia (17) but the painter is concerned with the scene as a whole, especially with the two players. In some the pieces are not readily identifiable, or are only named.(18) However, two Shahnama manuscripts in Berlin, one dated 1489 and the other 1593, include miniatures in which it is possible to recognize a few of them. (19) These all correspond to style set B, including the Bishop and Knight.

Figure 2 (left) Figure of an elephant, ivory, possibly a chess piece. Iraq, late 9th-early l0th century AD. Florence, Museo Nazionale del Bargello, inv. no.63C.

Figure 3 (right) Chariot (Rook), ivory. Reportedly from the Samarkand area, 7th century (?) AD. London, British Museum, inv. no. OA 1991.10-12.1.

(p. 113)

Figure 4 (top) Ivory chess pieces excavated at Nishapur, 9th century AD. New York, Metropolitan Museum of Art, inv. no.40.170.148.151 (after Wilkinson (1943)).
Style set A.

Figure 5 (bottom) Knight, stone. 9th century AD. Kuwait, Dar al-Athar al-Islamiyya, inv. no.50 Sb (photograph courtesy of the Al-Sabah Collection, Dar al-Athar al-Islamiyya, Kuwait National Museum). Style set A.

(p. 114)

Figure 6 Knight, ceramic, c.11th century AD. Kuwait, Dar al-Athar al-Islamiyya, inv. no. LNS 2730 (photograph courtesy of the Al-Sabah Collection, Dar al-Athar al-Islamiyya, Kuwait National Museum). Style set A.

The Ashmolean pieces may be approached initially through comparison with other Islamic abstract examples which, in general, have been identified by their resemblance either to modern Islamic abstract chessmen or to their earlier European counterparts. It is helpful, as a first step, to consider each piece in context, that is, to assign it to a particular style set and relate it to pieces in other col lections. Two sets may be distinguished; a list of the pieces belonging to each is given in Appendix I.